Methods for performance monitoring of channel estimation for demodulation

Abstract

A wireless transmit / receive unit (WTRU) may receive, from a network, training data information related to a model. The training data information comprises information related to demodulation reference signal (DMRS) resource elements (REs) that were used to train the model. The WTRU may determine that the model has drifted based on the set of preprocessed DMRS, the results of an out-of-distribution (OOD) detection test performed on the set of preprocessed DMRS REs, and the training data information. The WTRU may send an indication that the model has drifted to the network device. The indication that the model has drifted comprises results of a model drift detection test performed by the WTRU. The results of the OOD detection test comprise a binary result of the OOD detection test. The training data information comprises statistics associated with the preprocessed received DMRS REs that were used to train the model.

Description

METHODS FOR PERFORMANCE MONITORING OF CHANNEL ESTIMATION FOR DEMODULATIONCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 610,836, filed on December 15, 2023, the contents of which are hereby incorporated by references herein.BACKGROUND

[0002] Artificial intelligence (Al) is broadly defined as the behavior exhibited by machines that mimic cognitive functions to sense, reason, adapt and act. An Al component refers to the realization of behaviors and / or conformance to requirements by learning based on data, without explicit configuration of sequence of steps of actions. Such Al component may enable learning complex behaviors which is difficult to specify and / or implement when using legacy methods.

[0003] Machine learning (ML) refers to the type of algorithms that solve a problem based on learning through experience (‘data’), without being explicitly programmed (‘configuring a set of rules’). ML can be considered as a subset of Al. Different ML paradigms may be envisioned based on the nature of data or feedback available to the learning algorithm.

[0004] Deep learning refers to a class of ML algorithms that employ artificial neural networks, specifically, deep neural networks (DNNs), which were loosely inspired from biological systems. The DNNs are a special class of ML models that are inspired by the human brain wherein the input is linearly transformed and pass through non-linear activation function multiple times. DNNs consist typically of multiple layers where each layer consists of linear transformation and a given non-linear activation functions.SUMMARY

[0005] When an artificial intelligence machine learning-based (AI / ML-based) channel estimator for demodulation is trained (e.g., either online or offline), its performance may degrade over time due to model drift. The new dataset during the inference might havea different probability distribution than the dataset used during the training. Hence, using the original AI / ML model with the new data distribution may cause a performance degradation. Then, the performance of the AI / ML model cannot be guaranteed and / or has some uncertainty in generalization.

[0006] According to one exemplary aspect, the disclosure relates to a wireless transmit / receive unit (WTRU) that includes a processor and a memory, that is configured to receive, from a network device, training data information related to a model. The training data information may include information related to demodulation reference signal (DMRS) resource elements (REs) that were used to train the model. The WTRU may be configured to determine that the model has drifted based on a set of preprocessed received DMRS REs, the results of an out-of-distribution (OOD) detection test performed on the set of preprocessed received DMRS REs, and the training data information. The WTRU may be configured to send an indication that the model has drifted to the network device.

[0007] In an example, the indication that the model has drifted comprises results of a model drift detection test performed by the WTRU. In an example, the results of the OOD detection tests comprise a binary result of the OOD detection test. In an example, the training data information comprises statistics (e.g., cumulative density function (CDF), statistical mean, standard deviation, statistical moments, etc.) associated with the preprocessed received DMRS REs that were used to train the model. In an example, the processor is configured to receive a set of DMRS REs and preprocess the set of received DMRS REs to generate the set of preprocessed received DMRS REs. In an example, the processor is configured to determine that the model has drifted further based on a comparison of an output of a drift detection test to a threshold. The threshold is received from the network device. In an example, the processor is configured to send a request to the network device to receive updated information for the model. In an example, the processor is configured to send a request the network device to receive updated information for the model, and use a fallback algorithm for channel estimation for demodulation until the WTRU receives the updated information for the model. In an example, the processor is configured to receive updated information to update the model, wherein the indication that the model has drifted comprises resultsof the model drift detection algorithm. In an example, the processor is configured to determine that the model has drifted using a drift detection test, wherein the drift detection test comprises a Kolmogorov-Smirnov test, a Kullback-Leibler divergence test, a Jensen-Shannon distance test, or a Page-Hinkley test.

[0008] According to one exemplary aspect, the disclosure relates to a method implemented by a WTRU. The method comprising receiving, from a network device, training data information related to a model. The training data information may include information related to DMRS REs that were used to train the model. The method further comprising determining that the model has drifted based on a set of preprocessed received DMRS REs, the results of an out-of-distribution (OOD) detection test performed on the set of preprocessed received DMRS REs, and the training data information. The method further comprises sending an indication that the model has drifted to the network device.

[0009] In an example, the indication that the model has drifted comprises results of a model drift detection test performed by the WTRU. In an example, the results of the OOD detection tests comprise a binary result of the OOD detection test. In an example, the training data information comprises statistics (e.g., CDF, statistical mean, standard deviation, statistical moments, etc.) associated with the preprocessed received DMRS REs that were used to train the model. In an example, the method further comprises receiving a set of DMRS REs and preprocessing the set of receive DMRS REs to generate the set of preprocessed received DMRS REs. In an example, the method further comprises determining that the model has drifted further based on a comparison of an output of a drift detection test to a threshold. The threshold is received from the network device. In an example, the method further comprises sending a request to the network device to receive updated information for the model. In an example, the method further comprises sending a request the network device to receive updated information for the model and use a fallback algorithm for channel estimation for demodulation until the WTRU receives the updated information for the model. In an example, the method further comprises receiving updated information to update the model, wherein the indication that the model has drifted comprises results of the model drift detection algorithm. In an example, the method further comprises determining that the model hasdrifted using a drift detection test, wherein the drift detection test comprises a Kolmogorov-Smirnov test, a Kullback-Leibler divergence test, a Jensen-Shannon distance test, or a Page-Hinkley test.

[0010] The WTRLI may estimate an effective channel at each RE associated to the preprocessed received DMRS. If the OOD detection test result is the binary output, and if the binary output is zero, the WTRU may use an AI / ML-based channel estimator to predict the effective channel at the each RE associated to the preprocessed received DMRS. If the OOD detection test result is the binary output, and if the binary output is one, the WTRU may use a fallback algorithm-based channel estimator to predict the effective channel at each RE associated to the preprocessed received DMRS. If the OOD detection test result is a floating-point number between zero and one that approximates a probability of OOD, and if the floating-point number is above a threshold, the WTRU may use a fallback algorithm-based channel estimator to predict the effective channel at the each RE associated to the preprocessed received DMRS. If the OOD detection test result is a floating-point number between zero and one that approximates a probability of OOD, and if the floating-point number is below a threshold, the WTRU may use an AI / ML-based channel estimator to predict the effective channel at the each RE associated to the preprocessed received DMRS. A WTRU may receive training data information from a gNodeB (gNB). The WTRU may store preprocessed received DMRS REs within a first set with a preconfigured size. The WTRU may store the result of the OOD detection test within a second set with a preconfigured size. The WTRU may determine whether the WTRU is configured to send the result of the OOD detection test to the gNB.

[0011] If the WTRU is configured to send the result of the OOD detection to the gNB; the WTRU may determine whether an AI / ML model is drifting. If the AI / ML model is drifting, the WTRU may determine whether a total drift is above a first threshold. For example, if the total drift is above the first threshold and is below a second threshold, the WTRU may be informed by the gNB to keep using the AI / ML model, and / or the WTRU may request the gNB to update / retrain the AI / ML model online. For example, the WTRU may send a request to the network device to receive updated information for the model. For another example, if the total drift is above the first threshold and is above asecond threshold, the WTRU may request the gNB to update / retrain the AI / ML model online; and / or the WTRU may be requested by the gNB to use a fallback algorithm for channel estimation for demodulation. For another example, if the AI / ML model is not drifting; the WTRU may be informed by the gNB to keep using the AI / ML model.

[0012] If the WTRU is not configured to send the result of the OOD detection to the gNB, the WTRU may determine whether an AI / ML model is drifting. If the AI / ML model is drifting, the WTRU may determine whether a total drift is above a first threshold. For example, if the total drift is above the first threshold and is below a second threshold, the WTRU may use the AI / ML model, and / or the WTRU may request updating / retraining of the AI / ML model online. For example, the WTRU may send a request to the network device to receive updated information for the model. For another example, if the total drift is above the first threshold and is above a second threshold, the WTRU may request updating / retraining of the AI / ML model online, and / or the WTRU may use a fallback algorithm for channel estimation for demodulation. For another example, if the AI / ML model is not drifting, the WTRU may use the AI / ML model.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] FIG. 1A is a system diagram illustrating an example communications system in which one or more disclosed embodiments may be implemented.

[0014] FIG. 1 B is a system diagram illustrating an example wireless transmit / receive unit (WTRU) that may be used within the communications system illustrated in FIG. 1A according to an embodiment.

[0015] FIG. 1 C is a system diagram illustrating an example radio access network (RAN) and an example core network (CN) that may be used within the communications system illustrated in FIG. 1A according to an embodiment.

[0016] FIG. 1 D is a system diagram illustrating a further example RAN and a further example CN that may be used within the communications system illustrated in FIG. 1A according to an embodiment. 2

[0017] FIG. 2 illustrates an example of new radio (NR) demodulation reference signal (DMRS) symbol configuration with Type 1 for Ports 1000-1003.

[0018] FIG. 3 illustrates an example of NR DMRS symbol configuration with Type 2 for Ports 1000-1003.

[0019] FIG. 4 illustrates a flow diagram of an example of DMRS-based channel estimation, equalization, demodulation, and channel decoding.

[0020] FIG. 5 illustrates an example of a ReEsNet model.

[0021] FIG. 6 illustrates a flowchart of example procedures for the WTRU to perform an out-of-distribution (OOD) detection test in channel estimation for demodulation.

[0022] FIG. 7 illustrates an example of a DMRS OOD detection mechanism in the WTRU.

[0023] FIG. 8A-8C illustrate a flowchart of examples procedures for the WTRU to perform artificial intelligence (Al) / machine learning (ML) model drift detection in channel estimation for demodulation.

[0024] FIG. 9 illustrates an example of an AI / ML model drift detection mechanism in the WTRU.DETAILED DESCRIPTION

[0025] FIG. 1A is a diagram illustrating an example communications system 100 in which one or more disclosed embodiments may be implemented. The communications system 100 may be a multiple access system that provides content, such as voice, data, video, messaging, broadcast, etc., to multiple wireless users. The communications system 100 may enable multiple wireless users to access such content through the sharing of system resources, including wireless bandwidth. For example, the communications systems 100 may employ one or more channel access methods, such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal FDMA (OFDMA), single-carrier FDMA (SC-FDMA), zero-tail unique-word DFT-Spread OFDM (ZT UW DTS-s OFDM), unique word OFDM (UW-OFDM), resource block-filtered OFDM, filter bank multicarrier (FBMC), and the like.

[0026] As shown in FIG. 1A, the communications system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, a RAN 104 / 113, a CN 106 / 115, a public switched telephone network (PSTN) 108, the Internet 110, and othernetworks 112, though it will be appreciated that the disclosed embodiments contemplate any number of WTRUs, base stations, networks, and / or network elements. Each of the WTRUs 102a, 102b, 102c, 102d may be any type of device configured to operate and / or communicate in a wireless environment. By way of example, the WTRUs 102a, 102b, 102c, 102d, any of which may be referred to as a “station” and / or a “STA”, may be configured to transmit and / or receive wireless signals and may include a user equipment (UE), a mobile station, a fixed or mobile subscriber unit, a subscriptionbased unit, a pager, a cellular telephone, a personal digital assistant (PDA), a smartphone, a laptop, a netbook, a personal computer, a wireless sensor, a hotspot or Mi-Fi device, an Internet of Things (loT) device, a watch or other wearable, a headmounted display (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like. Any of the WTRUs 102a, 102b, 102c and 102d may be interchangeably referred to as a WTRU.

[0027] The communications systems 100 may also include a base station 114a and / or a base station 114b. Each of the base stations 114a, 114b may be any type of device configured to wirelessly interface with at least one of the WTRUs 102a, 102b, 102c, 102d to facilitate access to one or more communication networks, such as the CN 106 / 115, the Internet 110, and / or the other networks 112. By way of example, the base stations 114a, 114b may be a base transceiver station (BTS), a Node-B, an eNode B, a Home Node B, a Home eNode B, a gNB, a NR NodeB, a site controller, an access point (AP), a wireless router, and the like. While the base stations 114a, 114b are each depicted as a single element, it will be appreciated that the base stations 114a, 114b may include any number of interconnected base stations and / or network elements.

[0028] The base station 114a may be part of the RAN 104 / 113, which may also include other base stations and / or network elements (not shown), such as a base station controller (BSC), a radio network controller (RNC), relay nodes, etc. The base station 114a and / or the base station 114b may be configured to transmit and / or receive wireless signals on one or more carrier frequencies, which may be referred to as a cell(not shown). These frequencies may be in licensed spectrum, unlicensed spectrum, or a combination of licensed and unlicensed spectrum. A cell may provide coverage for a wireless service to a specific geographical area that may be relatively fixed or that may change over time. The cell may further be divided into cell sectors. For example, the cell associated with the base station 114a may be divided into three sectors. Thus, in one embodiment, the base station 114a may include three transceivers, i.e. , one for each sector of the cell. In an embodiment, the base station 114a may employ multiple-input multiple output (MIMO) technology and may utilize multiple transceivers for each sector of the cell. For example, beamforming may be used to transmit and / or receive signals in desired spatial directions.

[0029] The base stations 114a, 114b may communicate with one or more of the WTRUs 102a, 102b, 102c, 102d over an air interface 116, which may be any suitable wireless communication link (e.g., radio frequency (RF), microwave, centimeter wave, micrometer wave, infrared (IR), ultraviolet (UV), visible light, etc.). The air interface 116 may be established using any suitable radio access technology (RAT).

[0030] More specifically, as noted above, the communications system 100 may be a multiple access system and may employ one or more channel access schemes, such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA, and the like. For example, the base station 114a in the RAN 104 / 113 and the WTRUs 102a, 102b, 102c may implement a radio technology such as Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access (UTRA), which may establish the air interface 115 / 116 / 117 using wideband CDMA (WCDMA). WCDMA may include communication protocols such as High-Speed Packet Access (HSPA) and / or Evolved HSPA (HSPA+). HSPA may include High-Speed Downlink (DL) Packet Access (HSDPA) and / or High-Speed UL Packet Access (HSUPA).

[0031] In an embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement a radio technology such as Evolved UMTS Terrestrial Radio Access (E- UTRA), which may establish the air interface 116 using Long Term Evolution (LTE) and / or LTE-Advanced (LTE-A) and / or LTE-Advanced Pro (LTE -A Pro).

[0032] In an embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement a radio technology such as NR Radio Access, which may establish the air interface 116 using New Radio (NR).

[0033] In an embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement multiple radio access technologies. For example, the base station 114a and the WTRUs 102a, 102b, 102c may implement LTE radio access and NR radio access together, for instance using dual connectivity (DC) principles. Thus, the air interface utilized by WTRUs 102a, 102b, 102c may be characterized by multiple types of radio access technologies and / or transmissions sent to / from multiple types of base stations (e.g., a eNB and a gNB).

[0034] In other embodiments, the base station 114a and the WTRUs 102a, 102b, 102c may implement radio technologies such as IEEE 802.11 (i.e., Wireless Fidelity (WiFi), IEEE 802.16 (i.e., Worldwide Interoperability for Microwave Access (WiMAX)), CDMA2000, CDMA2000 1X, CDMA2000 EV-DO, Interim Standard 2000 (IS-2000), Interim Standard 95 (IS-95), Interim Standard 856 (IS-856), Global System for Mobile communications (GSM), Enhanced Data rates for GSM Evolution (EDGE), GSM EDGE (GERAN), and the like.

[0035] The base station 114b in FIG. 1A may be a wireless router, Home Node B, Home eNode B, or access point, for example, and may utilize any suitable RAT for facilitating wireless connectivity in a localized area, such as a place of business, a home, a vehicle, a campus, an industrial facility, an air corridor (e.g., for use by drones), a roadway, and the like. In one embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.11 to establish a wireless local area network (WLAN). In an embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.15 to establish a wireless personal area network (WPAN). In yet another embodiment, the base station 114b and the WTRUs 102c, 102d may utilize a cellular-based RAT (e.g., WCDMA, CDMA2000, GSM, LTE, LTE-A, LTE-A Pro, NR etc.) to establish a picocell or femtocell. As shown in FIG. 1 A, the base station 114b may have a direct connection to the Internet 110. Thus, the base station 114b may not be required to access the Internet 110 via the CN 106 / 115.

[0036] The RAN 104 / 113 may be in communication with the CN 106 / 115, which may be any type of network configured to provide voice, data, applications, and / or voice over internet protocol (VoIP) services to one or more of the WTRUs 102a, 102b, 102c, 102d. The data may have varying quality of service (QoS) requirements, such as differing throughput requirements, latency requirements, error tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, and the like. The CN 106 / 115 may provide call control, billing services, mobile location-based services, pre-paid calling, Internet connectivity, video distribution, etc., and / or perform high-level security functions, such as user authentication. Although not shown in FIG. 1A, it will be appreciated that the RAN 104 / 113 and / or the CN 106 / 115 may be in direct or indirect communication with other RANs that employ the same RAT as the RAN 104 / 113 or a different RAT. For example, in addition to being connected to the RAN 104 / 113, which may be utilizing a NR radio technology, the CN 106 / 115 may also be in communication with another RAN (not shown) employing a GSM, UMTS, CDMA 2000, WiMAX, E- UTRA, or WiFi radio technology.

[0037] The CN 106 / 115 may also serve as a gateway for the WTRUs 102a, 102b, 102c, 102d to access the PSTN 108, the Internet 110, and / or the other networks 112. The PSTN 108 may include circuit-switched telephone networks that provide plain old telephone service (POTS). The Internet 110 may include a global system of interconnected computer networks and devices that use common communication protocols, such as the transmission control protocol (TCP), user datagram protocol (UDP) and / or the internet protocol (IP) in the TCP / IP internet protocol suite. The networks 112 may include wired and / or wireless communications networks owned and / or operated by other service providers. For example, the networks 112 may include another CN connected to one or more RANs, which may employ the same RAT as the RAN 104 / 113 or a different RAT.

[0038] Some or all of the WTRUs 102a, 102b, 102c, 102d in the communications system 100 may include multi-mode capabilities (e.g., the WTRUs 102a, 102b, 102c, 102d may include multiple transceivers for communicating with different wireless networks over different wireless links). For example, the WTRU 102c shown in FIG. 1 A may be configured to communicate with the base station 114a, which may employ acellular-based radio technology, and with the base station 114b, which may employ an IEEE 802 radio technology.

[0039] FIG. 1 B is a system diagram illustrating an example WTRU 102. As shown in FIG. 1 B, the WTRU 102 may include a processor 118, a transceiver 120, a transmit / receive element 122, a speaker / microphone 124, a keypad 126, a display / touchpad 128, non-removable memory 130, removable memory 132, a power source 134, a global positioning system (GPS) chipset 136, and / or other peripherals 138, among others. It will be appreciated that the WTRU 102 may include any subcombination of the foregoing elements while remaining consistent with an embodiment.

[0040] The processor 118 may be a general purpose processor, a special purpose processor, a conventional processor, a digital signal processor (DSP), a plurality of microprocessors, one or more microprocessors in association with a DSP core, a controller, a microcontroller, Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) circuits, any other type of integrated circuit (IC), a state machine, and the like. The processor 118 may perform signal coding, data processing, power control, input / output processing, and / or any other functionality that enables the WTRU 102 to operate in a wireless environment. The processor 118 may be coupled to the transceiver 120, which may be coupled to the transmit / receive element 122. While FIG. 1 B depicts the processor 118 and the transceiver 120 as separate components, it will be appreciated that the processor 118 and the transceiver 120 may be integrated together in an electronic package or chip.

[0041] The transmit / receive element 122 may be configured to transmit signals to, or receive signals from, a base station (e g., the base station 114a) over the air interface 116. For example, in one embodiment, the transmit / receive element 122 may be an antenna configured to transmit and / or receive RF signals. In an embodiment, the transmit / receive element 122 may be an emitter / detector configured to transmit and / or receive IR, UV, or visible light signals, for example. In yet another embodiment, the transmit / receive element 122 may be configured to transmit and / or receive both RF and light signals. It will be appreciated that the transmit / receive element 122 may be configured to transmit and / or receive any combination of wireless signals.

[0042] Although the transmit / receive element 122 is depicted in FIG. 1 B as a single element, the WTRU 102 may include any number of transmit / receive elements 122. More specifically, the WTRU 102 may employ MIMO technology. Thus, in one embodiment, the WTRU 102 may include two or more transmit / receive elements 122 (e.g., multiple antennas) for transmitting and receiving wireless signals over the air interface 116.

[0043] The transceiver 120 may be configured to modulate the signals that are to be transmitted by the transmit / receive element 122 and to demodulate the signals that are received by the transmit / receive element 122. As noted above, the WTRU 102 may have multi-mode capabilities. Thus, the transceiver 120 may include multiple transceivers for enabling the WTRU 102 to communicate via multiple RATs, such as NR and IEEE 802.11 , for example.

[0044] The processor 118 of the WTRU 102 may be coupled to, and may receive user input data from, the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128 (e.g., a liquid crystal display (LCD) display unit or organic lightemitting diode (OLED) display unit). The processor 118 may also output user data to the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128. In addition, the processor 118 may access information from, and store data in, any type of suitable memory, such as the non-removable memory 130 and / or the removable memory 132. The non-removable memory 130 may include random-access memory (RAM), readonly memory (ROM), a hard disk, or any other type of memory storage device. The removable memory 132 may include a subscriber identity module (SIM) card, a memory stick, a secure digital (SD) memory card, and the like. In other embodiments, the processor 118 may access information from, and store data in, memory that is not physically located on the WTRU 102, such as on a server or a home computer (not shown).

[0045] The processor 118 may receive power from the power source 134, and may be configured to distribute and / or control the power to the other components in the WTRU 102. The power source 134 may be any suitable device for powering the WTRU 102. For example, the power source 134 may include one or more dry cell batteries (e.g.,nickel-cadmium (NiCd), nickel-zinc (NiZn), nickel metal hydride (NiMH), lithium-ion (Li- ion), etc.), solar cells, fuel cells, and the like.

[0046] The processor 118 may also be coupled to the GPS chipset 136, which may be configured to provide location information (e.g., longitude and latitude) regarding the current location of the WTRU 102. In addition to, or in lieu of, the information from the GPS chipset 136, the WTRU 102 may receive location information over the air interface 116 from a base station (e.g., base stations 114a, 114b) and / or determine its location based on the timing of the signals being received from two or more nearby base stations. It will be appreciated that the WTRU 102 may acquire location information by way of any suitable location-determination method while remaining consistent with an embodiment.

[0047] The processor 118 may further be coupled to other peripherals 138, which may include one or more software and / or hardware modules that provide additional features, functionality and / or wired or wireless connectivity. For example, the peripherals 138 may include an accelerometer, an e-compass, a satellite transceiver, a digital camera (for photographs and / or video), a universal serial bus (USB) port, a vibration device, a television transceiver, a hands free headset, a Bluetooth® module, a frequency modulated (FM) radio unit, a digital music player, a media player, a video game player module, an Internet browser, a Virtual Reality and / or Augmented Reality (VR / AR) device, an activity tracker, and the like. The peripherals 138 may include one or more sensors, the sensors may be one or more of a gyroscope, an accelerometer, a hall effect sensor, a magnetometer, an orientation sensor, a proximity sensor, a temperature sensor, a time sensor; a geolocation sensor; an altimeter, a light sensor, a touch sensor, a magnetometer, a barometer, a gesture sensor, a biometric sensor, and / or a humidity sensor.

[0048] The WTRU 102 may include a full duplex radio for which transmission and reception of some or all of the signals (e.g., associated with particular subframes for both the UL (e.g., for transmission) and downlink (e.g., for reception) may be concurrent and / or simultaneous. The full duplex radio may include an interference management unit 139 to reduce and or substantially eliminate self-interference via either hardware (e.g., a choke) or signal processing via a processor (e.g., a separate processor (notshown) or via processor 118). In an embodiment, the WRTU 102 may include a halfduplex radio for which transmission and reception of some or all of the signals (e.g., associated with particular subframes for either the UL (e.g., for transmission) or the downlink (e.g., for reception)).

[0049] FIG. 1 C is a system diagram illustrating the RAN 104 and the CN 106 according to an embodiment. As noted above, the RAN 104 may employ an E-UTRA radio technology to communicate with the WTRUs 102a, 102b, 102c over the air interface 116. The RAN 104 may also be in communication with the CN 106.

[0050] The RAN 104 may include eNode-Bs 160a, 160b, 160c, though it will be appreciated that the RAN 104 may include any number of eNode-Bs while remaining consistent with an embodiment. The eNode-Bs 160a, 160b, 160c may each include one or more transceivers for communicating with the WTRUs 102a, 102b, 102c over the air interface 116. In one embodiment, the eNode-Bs 160a, 160b, 160c may implement MIMO technology. Thus, the eNode-B 160a, for example, may use multiple antennas to transmit wireless signals to, and / or receive wireless signals from, the WTRU 102a.

[0051] Each of the eNode-Bs 160a, 160b, 160c may be associated with a particular cell (not shown) and may be configured to handle radio resource management decisions, handover decisions, scheduling of users in the UL and / or DL, and the like. As shown in FIG. 1 C, the eNode-Bs 160a, 160b, 160c may communicate with one another over an X2 interface.

[0052] The CN 106 shown in FIG. 1 C may include a mobility management entity (MME) 162, a serving gateway (SGW) 164, and a packet data network (PDN) gateway (or PGW) 166. While each of the foregoing elements are depicted as part of the CN 106, it will be appreciated that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0053] The MME 162 may be connected to each of the eNode-Bs 162a, 162b, 162c in the RAN 104 via an S1 interface and may serve as a control node. For example, the MME 162 may be responsible for authenticating users of the WTRUs 102a, 102b, 102c, bearer activation / deactivation, selecting a particular serving gateway during an initial attach of the WTRUs 102a, 102b, 102c, and the like. The MME 162 may provide acontrol plane function for switching between the RAN 104 and other RANs (not shown) that employ other radio technologies, such as GSM and / or WCDMA.

[0054] The SGW 164 may be connected to each of the eNode Bs 160a, 160b, 160c in the RAN 104 via the S1 interface. The SGW 164 may generally route and forward user data packets to / from the WTRUs 102a, 102b, 102c. The SGW 164 may perform other functions, such as anchoring user planes during inter-eNode B handovers, triggering paging when DL data is available for the WTRUs 102a, 102b, 102c, managing and storing contexts of the WTRUs 102a, 102b, 102c, and the like.

[0055] The SGW 164 may be connected to the PGW 166, which may provide the WTRUs 102a, 102b, 102c with access to packet-switched networks, such as the Internet 110, to facilitate communications between the WTRUs 102a, 102b, 102c and IP-enabled devices.

[0056] The ON 106 may facilitate communications with other networks. For example, the ON 106 may provide the WTRUs 102a, 102b, 102c with access to circuit-switched networks, such as the PSTN 108, to facilitate communications between the WTRUs 102a, 102b, 102c and traditional land-line communications devices. For example, the CN 106 may include, or may communicate with, an IP gateway (e.g., an IP multimedia subsystem (IMS) server) that serves as an interface between the CN 106 and the PSTN 108. In addition, the CN 106 may provide the WTRUs 102a, 102b, 102c with access to the other networks 112, which may include other wired and / or wireless networks that are owned and / or operated by other service providers.

[0057] Although the WTRU is described in FIGS. 1 A-1 D as a wireless terminal, it is contemplated that in certain representative embodiments that such a terminal may use (e.g., temporarily or permanently) wired communication interfaces with the communication network.

[0058] In representative embodiments, the other network 112 may be a WLAN.

[0059] A WLAN in Infrastructure Basic Service Set (BSS) mode may have an Access Point (AP) for the BSS and one or more stations (STAs) associated with the AP. The AP may have an access or an interface to a Distribution System (DS) or another type of wired / wireless network that carries traffic in to and / or out of the BSS. Traffic to STAs that originates from outside the BSS may arrive through the AP and may be delivered tothe STAs. Traffic originating from STAs to destinations outside the BSS may be sent to the AP to be delivered to respective destinations. Traffic between STAs within the BSS may be sent through the AP, for example, where the source STA may send traffic to the AP and the AP may deliver the traffic to the destination STA. The traffic between STAs within a BSS may be considered and / or referred to as peer-to-peer traffic. The peer-to- peer traffic may be sent between (e.g., directly between) the source and destination STAs with a direct link setup (DLS). In certain representative embodiments, the DLS may use an 802.11 e DLS or an 802.11 z tunneled DLS (TDLS). A WLAN using an Independent BSS (IBSS) mode may not have an AP, and the STAs (e.g., all of the STAs) within or using the IBSS may communicate directly with each other. The IBSS mode of communication may sometimes be referred to herein as an “ad-hoc” mode of communication.

[0060] When using the 802.11ac infrastructure mode of operation or a similar mode of operations, the AP may transmit a beacon on a fixed channel, such as a primary channel. The primary channel may be a fixed width (e.g., 20 MHz wide bandwidth) or a dynamically set width via signaling. The primary channel may be the operating channel of the BSS and may be used by the STAs to establish a connection with the AP. In certain representative embodiments, Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA) may be implemented, for example in in 802.11 systems. For CSMA / CA, the STAs (e.g., every STA), including the AP, may sense the primary channel. If the primary channel is sensed / detected and / or determined to be busy by a particular STA, the particular STA may back off. One STA (e.g., only one station) may transmit at any given time in a given BSS.

[0061] High Throughput (HT) STAs may use a 40 MHz wide channel for communication, for example, via a combination of the primary 20 MHz channel with an adjacent or nonadjacent 20 MHz channel to form a 40 MHz wide channel.

[0062] Very High Throughput (VHT) STAs may support 20MHz, 40 MHz, 80 MHz, and / or 160 MHz wide channels. The 40 MHz, and / or 80 MHz, channels may be formed by combining contiguous 20 MHz channels. A 160 MHz channel may be formed by combining 8 contiguous 20 MHz channels, or by combining two non-contiguous 80 MHz channels, which may be referred to as an 80+80 configuration. For the 80+80configuration, the data, after channel encoding, may be passed through a segment parser that may divide the data into two streams. Inverse Fast Fourier Transform (IFFT) processing, and time domain processing, may be done on each stream separately. The streams may be mapped on to the two 80 MHz channels, and the data may be transmitted by a transmitting STA. At the receiver of the receiving STA, the above described operation for the 80+80 configuration may be reversed, and the combined data may be sent to the Medium Access Control (MAC).

[0063] Sub 1 GHz modes of operation are supported by 802.11 af and 802.11 ah. The channel operating bandwidths, and carriers, are reduced in 802.11af and 802.11 ah relative to those used in 802.11 n, and 802.11ac. 802.11 af supports 5 MHz, 10 MHz and 20 MHz bandwidths in the TV White Space (TVWS) spectrum, and 802.11 ah supports 1 MHz, 2 MHz, 4 MHz, 8 MHz, and 16 MHz bandwidths using non-TVWS spectrum. According to a representative embodiment, 802.11 ah may support Meter Type Control / Machine-Type Communications, such as MTC devices in a macro coverage area. MTC devices may have certain capabilities, for example, limited capabilities including support for (e.g., only support for) certain and / or limited bandwidths. The MTC devices may include a battery with a battery life above a threshold (e.g., to maintain a very long battery life).

[0064] WLAN systems, which may support multiple channels, and channel bandwidths, such as 802.11 n, 802.11 ac, 802.11 af, and 802.11 ah, include a channel which may be designated as the primary channel. The primary channel may have a bandwidth equal to the largest common operating bandwidth supported by all STAs in the BSS. The bandwidth of the primary channel may be set and / or limited by a STA, from among all STAs in operating in a BSS, which supports the smallest bandwidth operating mode. In the example of 802.11 ah, the primary channel may be 1 MHz wide for STAs (e.g., MTC type devices) that support (e.g., only support) a 1 MHz mode, even if the AP, and other STAs in the BSS support 2 MHz, 4 MHz, 8 MHz, 16 MHz, and / or other channel bandwidth operating modes. Carrier sensing and / or Network Allocation Vector (NAV) settings may depend on the status of the primary channel. If the primary channel is busy, for example, due to a STA (which supports only a 1 MHz operating mode),transmitting to the AP, the entire available frequency bands may be considered busy even though a majority of the frequency bands remains idle and may be available.

[0065] In the United States, the available frequency bands, which may be used by 802.11 ah, are from 902 MHz to 928 MHz. In Korea, the available frequency bands are from 917.5 MHz to 923.5 MHz. In Japan, the available frequency bands are from 916.5 MHz to 927.5 MHz. The total bandwidth available for 802.11 ah is 6 MHz to 26 MHz depending on the country code.

[0066] FIG. 1 D is a system diagram illustrating the RAN 113 and the CN 115 according to an embodiment. As noted above, the RAN 113 may employ an NR radio technology to communicate with the WTRUs 102a, 102b, 102c over the air interface 116. The RAN 113 may also be in communication with the CN 115.

[0067] The RAN 113 may include gNBs 180a, 180b, 180c, though it will be appreciated that the RAN 113 may include any number of gNBs while remaining consistent with an embodiment. The gNBs 180a, 180b, 180c may each include one or more transceivers for communicating with the WTRUs 102a, 102b, 102c over the air interface 116. In one embodiment, the gNBs 180a, 180b, 180c may implement MIMO technology. For example, gNBs 180a, 108b may utilize beamforming to transmit signals to and / or receive signals from the gNBs 180a, 180b, 180c. Thus, the gNB 180a, for example, may use multiple antennas to transmit wireless signals to, and / or receive wireless signals from, the WTRU 102a. In an embodiment, the gNBs 180a, 180b, 180c may implement carrier aggregation technology. For example, the gNB 180a may transmit multiple component carriers to the WTRU 102a (not shown). A subset of these component carriers may be on unlicensed spectrum while the remaining component carriers may be on licensed spectrum. In an embodiment, the gNBs 180a, 180b, 180c may implement Coordinated Multi-Point (CoMP) technology. For example, WTRU 102a may receive coordinated transmissions from gNB 180a and gNB 180b (and / or gNB 180c).

[0068] The WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c using transmissions associated with a scalable numerology. For example, the OFDM symbol spacing and / or OFDM subcarrier spacing may vary for different transmissions, different cells, and / or different portions of the wireless transmission spectrum. The WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c usingsubframe or transmission time intervals (TTIs) of various or scalable lengths (e.g., containing varying number of OFDM symbols and / or lasting varying lengths of absolute time).

[0069] The gNBs 180a, 180b, 180c may be configured to communicate with the WTRUs 102a, 102b, 102c in a standalone configuration and / or a non-standalone configuration. In the standalone configuration, WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c without also accessing other RANs (e.g., such as eNode-Bs 160a, 160b, 160c). In the standalone configuration, WTRUs 102a, 102b, 102c may utilize one or more of gNBs 180a, 180b, 180c as a mobility anchor point. In the standalone configuration, WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c using signals in an unlicensed band. In a non-standalone configuration WTRUs 102a, 102b, 102c may communicate with / connect to gNBs 180a, 180b, 180c while also communicating with / connecting to another RAN such as eNode-Bs 160a, 160b, 160c. For example, WTRUs 102a, 102b, 102c may implement DC principles to communicate with one or more gNBs 180a, 180b, 180c and one or more eNode-Bs 160a, 160b, 160c substantially simultaneously. In the non-standalone configuration, eNode-Bs 160a, 160b, 160c may serve as a mobility anchor for WTRUs 102a, 102b, 102c and gNBs 180a, 180b, 180c may provide additional coverage and / or throughput for servicing WTRUs 102a, 102b, 102c.

[0070] Each of the gNBs 180a, 180b, 180c may be associated with a particular cell (not shown) and may be configured to handle radio resource management decisions, handover decisions, scheduling of users in the UL and / or DL, support of network slicing, dual connectivity, interworking between NR and E-UTRA, routing of user plane data towards User Plane Function (UPF) 184a, 184b, routing of control plane information towards Access and Mobility Management Function (AMF) 182a, 182b and the like. As shown in FIG. 1 D, the gNBs 180a, 180b, 180c may communicate with one another over an Xn interface.

[0071] The CN 115 shown in FIG. 1 D may include at least one AMF 182a, 182b, at least one UPF 184a, 184b, at least one Session Management Function (SMF) 183a, 183b, and possibly a Data Network (DN) 185a, 185b. While each of the foregoingelements are depicted as part of the CN 115, it will be appreciated that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0072] The AMF 182a, 182b may be connected to one or more of the gNBs 180a, 180b, 180c in the RAN 113 via an N2 interface and may serve as a control node. For example, the AMF 182a, 182b may be responsible for authenticating users of the WTRUs 102a, 102b, 102c, support for network slicing (e.g., handling of different PDU sessions with different requirements), selecting a particular SMF 183a, 183b, management of the registration area, termination of NAS signaling, mobility management, and the like. Network slicing may be used by the AMF 182a, 182b in order to customize CN support for WTRUs 102a, 102b, 102c based on the types of services being utilized WTRUs 102a, 102b, 102c. For example, different network slices may be established for different use cases such as services relying on ultra-reliable low latency (URLLC) access, services relying on enhanced massive mobile broadband (eMBB) access, services for machine type communication (MTC) access, and / or the like. The AMF 162 may provide a control plane function for switching between the RAN 113 and other RANs (not shown) that employ other radio technologies, such as LTE, LTE-A, LTE-A Pro, and / or non-3GPP access technologies such as WiFi.

[0073] The SMF 183a, 183b may be connected to an AMF 182a, 182b in the CN 115 via an N11 interface. The SMF 183a, 183b may also be connected to a UPF 184a, 184b in the CN 115 via an N4 interface. The SMF 183a, 183b may select and control the UPF 184a, 184b and configure the routing of traffic through the UPF 184a, 184b. The SMF 183a, 183b may perform other functions, such as managing and allocating WTRU IP address, managing PDU sessions, controlling policy enforcement and QoS, providing downlink data notifications, and the like. A PDU session type may be IP-based, non-IP based, Ethernet-based, and the like.

[0074] The UPF 184a, 184b may be connected to one or more of the gNBs 180a, 180b, 180c in the RAN 113 via an N3 interface, which may provide the WTRUs 102a, 102b, 102c with access to packet-switched networks, such as the Internet 110, to facilitate communications between the WTRUs 102a, 102b, 102c and IP-enabled devices. The UPF 184, 184b may perform other functions, such as routing and forwarding packets,enforcing user plane policies, supporting multi-homed PDU sessions, handling user plane QoS, buffering downlink packets, providing mobility anchoring, and the like.

[0075] The CN 115 may facilitate communications with other networks. For example, the CN 115 may include, or may communicate with, an IP gateway (e.g., an IP multimedia subsystem (IMS) server) that serves as an interface between the CN 115 and the PSTN 108. In addition, the CN 115 may provide the WTRUs 102a, 102b, 102c with access to the other networks 112, which may include other wired and / or wireless networks that are owned and / or operated by other service providers. In one embodiment, the WTRUs 102a, 102b, 102c may be connected to a local Data Network (DN) 185a, 185b through the UPF 184a, 184b via the N3 interface to the UPF 184a, 184b and an N6 interface between the UPF 184a, 184b and the DN 185a, 185b.

[0076] In view of Figures 1 A-1 D, and the corresponding description of Figures 1 A-1 D, one or more, or all, of the functions described herein with regard to one or more of: WTRU 102a-d, Base Station 114a-b, eNode-B 160a-c, MME 162, SGW 164, PGW 166, gNB 180a-c, AMF 182a-ab, UPF 184a-b, SMF 183a-b, DN 185a-b, and / or any other device(s) described herein, may be performed by one or more emulation devices (not shown). The emulation devices may be one or more devices configured to emulate one or more, or all, of the functions described herein. For example, the emulation devices may be used to test other devices and / or to simulate network and / or WTRU functions.

[0077] The emulation devices may be designed to implement one or more tests of other devices in a lab environment and / or in an operator network environment. For example, the one or more emulation devices may perform the one or more, or all, functions while being fully or partially implemented and / or deployed as part of a wired and / or wireless communication network in order to test other devices within the communication network. The one or more emulation devices may perform the one or more, or all, functions while being temporarily implemented / deployed as part of a wired and / or wireless communication network. The emulation device may be directly coupled to another device for purposes of testing and / or may performing testing using over-the-air wireless communications.

[0078] The one or more emulation devices may perform the one or more, including all, functions while not being implemented / deployed as part of a wired and / or wirelesscommunication network. For example, the emulation devices may be utilized in a testing scenario in a testing laboratory and / or a non-deployed (e.g., testing) wired and / or wireless communication network in order to implement testing of one or more components. The one or more emulation devices may be test equipment. Direct RF coupling and / or wireless communications via RF circuitry (e.g., which may include one or more antennas) may be used by the emulation devices to transmit and / or receive data.

[0079] When an artificial intelligence machine learning-based (AI / ML-based) channel estimator for demodulation is trained (e.g., either online or offline), its performance may degrade over time due to model drift. The new data during the inference may have a different probability distribution than the dataset used during the training. Hence, using the original AI / ML model with the new data distribution may cause a performance degradation. In addition, the performance of the AI / ML model may not be guaranteed and / or has some uncertainty in generalization.

[0080] Methods for performance monitoring of channel estimation for demodulation may enable a WTRU to perform out-of-distribution (OOD) detection to maintain a certain AI / ML model performance in channel estimation for demodulation. The methods may enable the WTRU to monitor the performance of AI / ML models utilized in channel estimation for demodulation. The methods may enable the WTRU to perform the AI / ML model drift detection in channel estimation for demodulation.

[0081] For systems using AI / ML models in channel estimation for demodulation, methods for a WTRU to monitor the performance of the AI / ML model may be described herein.

[0082] AI may be broadly defined as the behavior exhibited by machines that mimic cognitive functions to sense, reason, adapt and act. An Al component may refer to the realization of behaviors and / or conformance to requirements by learning based on data, without explicit configuration of sequence of steps of actions. Such Al component may enable learning complex behaviors which might be difficult to specify and / or implement when using legacy methods.

[0083] ML may refer to the type of algorithms that solve a problem based on learning through experience (‘data’), without being explicitly programmed (‘configuring a set ofrules’). ML may be considered as a subset of Al. Different ML paradigms may be envisioned based on the nature of data or feedback available to the learning algorithm. In one example, a supervised learning approach may involve learning a function that maps input to an output based on labeled training example. Each labeled training example may be a pair consisting of an input and its corresponding output. In one example, an unsupervised learning approach may involve detecting patterns in the data with no pre-existing labels. In one example, a reinforcement learning approach may involve performing sequence of actions in an environment to maximize the cumulative reward. It may be possible to apply ML algorithms using a combination or interpolation of the above-mentioned approaches. For example, a semi-supervised learning approach may use a combination of a small amount of labeled data with a large amount of unlabeled data during training. In this regard semi-supervised learning may fall between unsupervised learning (e.g., with no labeled training data) and supervised learning (e.g., with only labeled training data).

[0084] Deep learning may refer to a class of ML algorithms that employ artificial neural networks, specifically, deep neural networks (DNNs), which were loosely inspired from biological systems. The DNNs may be a special class of ML models that are inspired by the human brain wherein the input is linearly transformed and pass through non-linear activation function multiple times. DNNs may consist typically of multiple layers where each layer consists of linear transformation and a given non-linear activation functions. The DNNs may be trained using the training data via back-propagation algorithm. Recently, DNNs may have shown state-of-the-art performance in a variety of domains for various ML settings (e.g., supervised, un-supervised, semi-supervised, etc ). For example, the variety of domains may include speech, vision, natural language, wireless communication, and the like.

[0085] AI / ML models may demonstrate exceptional performance on controlled and experimental datasets. However, these models may operate with an underlying assumption. For example, the underlying assumption about the data encountered in real-world scenarios may conform to the same statistical distribution as the data used for training and testing. However, this assumption may not hold universally, and there may exist situations where the model’s performance will degrade.

[0086] This type of problem may be commonly known as OOD detection. The objective here may be to discern whether a new data sample, given a known dataset, adheres to the same underlying distribution or displays some form of atypical behavior. A fundamental assumption in OOD detection may be the scarcity of atypical samples. Consequently, OOD detection may be typically activated when atypical occurrences are diverse, infrequent, and their specific characteristics are not known until they manifest.

[0087] A variety of OOD detection algorithms may include but not be limited to classification, probability and generative models, typicality, and / or reconstruction. In classification OOD detection algorithm, the original dataset of in-distribution examples may be used to build a one-class classifier that identifies if a new example belongs to the same dataset. In the OOD detection algorithms of probability and generative models, a probability model of the in-distribution data may be built. New examples with low probability may be classified as out-of-distribution. In typical OOD detection algorithms, the algorithm may check whether statistics of the input example are typical of the in-distribution data. In reconstruction OOD detection algorithm, the algorithm may learn an autoencoder that compresses the in-distribution data to a low dimensional representation and then may reconstruct it from this representation. Out-of-distribution examples may be identified as examples where the reconstruction quality is low.

[0088] After deploying an AI / ML model for online inference, its performance may degrade over time due to a concept called “model drift.” The deployed AI / ML model may be constantly receiving new data to make predictions upon. However, this data may have a different probability distribution than the one used for training. Hence, using the original AI / ML model with the new data distribution may cause a drop in the model performance. In short, AI / ML model drift may be a situation where a model’s performance degrades over time, causing the model to start giving poor predictions. AI / ML model drift may be categorized into two categories: concept drift and data drift.

[0089] Concept drift may be the situation when the functional relationship between the inputs and outputs of an AI / ML models changes. In this situation, the context may have changed, but the AI / ML model may not know about the change. The AI / ML model’s learned parameters may not hold anymore. On the other hand, data drift may be the situation where the AI / ML model’s input distribution changes. The AI / ML model’sperformance may weaken because it receives data on which it has not been trained enough. The data drift may also occur if the data received by the AI / ML model at the online inference time contains features that were not present during the offline training.

[0090] Typically, training of AI / ML models may be based on data that is collected from real-field measurements, generated synthetically or artificially using some defined models, and / or a combination of real-world measurements augmented with synthetic data. Once the training is complete, the learnt parameters of the AI / ML model may stay fixed during the inference phase. This training methodology may imply that the data used for training is expected to be statistically similar to the data used for inference. Otherwise, the performance of the AI / ML model may not be guaranteed and / or has some uncertainty in generalization. Hence, this methodology may not perform well in the context of a practical system due to issues related to AI / ML model drift resulting from the statistical differences in the training and the inference data.

[0091] Coherent demodulation of signals transmitted over the radio interface may typically require knowledge of the effective (e.g., precoded) wireless channel. The channel estimation may process at the receiver in new radio (NR) relies on the transmission of physical channels accompanied with demodulation reference signals (DMRS). DMRSs may be generated using pseudo-random sequences based on systems parameters known to the receiver. The parameters used to control the sequence generation may include scrambling identity, symbol locations, number of orthogonal frequency-division multiplexing (OFDM) symbols in a slot, and the like. The DMRS operation in NR may include several predefined options for patterns (e.g., uniform / equally spaced patterns) and densities of RSs based on the physical channels, configured using scheduling (e.g., downlink control information (DCI)-based scheduling) and high-layer configuration to cater for different use cases and WTRU capabilities.

[0092] The configuration of the DMRS may include density and pattern in the resource grid, duration, starting symbol (e.g., front-loaded DMRS), and cover codes, to differentiate between antenna ports sharing the same time / frequency resources (for single-user and multi-user multiple input multiple output (MIMO) cases). The set of parameters for DMRS may be different depending on the physical channel and depending on WTRU capability, for example, for physical downlink shared channel(PDSCH) DMRS, the WTRU capabilities may include configuration type 1 or type 2, mapping type A or type B, starting symbol for mapping type A, single versus double symbol DMRS, DMRS additional positions, and duration.

[0093] The specific selection of DMRS may be carried out by both higher-layer configuration and dynamic (e.g., DCI-based) signaling, but also there may be cases where there is a default configuration in place. Upon selection of DMRS settings, the gNodeB (gNB) signals (e.g., gNB signals using radio resource control (RRC), MAC control element (MAC-CE), or PDCCH / DCI) may select the terminal. FIG. 2 illustrates an example of NR DMRS symbol configuration 200 with Type 1 for ports 1000-1003. FIG. 3 illustrates an example of NR DMRS symbol configuration 300 with type 2 for ports 1000-1003. FIG. 2 and FIG. 3 may present example DMRS patterns over one slot and one resource block in NR with code division multiplexing (CDM) grouping across the frequency and code domains. Specifically, in FIG. 2, DMRS pattern may follow mapping type A, configuration type 1 , and starting symbol 3 using downlink antenna ports 1000-1003. In FIG. 3, DMRS pattern may use mapping type A, configuration type 2 and starting symbol 2 using downlink antenna ports 1000-1003.

[0094] The terminal may utilize the DMRS for channel estimation and coherent demodulation of the corresponding physical channels. This may be achieved through specific receiver filter implementation (e.g., least squares, minimum mean squared error (MMSE), etc.) which broadly estimates the composite channel by mapping the transmitted layers onto the receive antennas for the resource blocks that are scheduled.

[0095] FIG. 4 illustrates a flow diagram of DMRS-based channel estimation, equalization, demodulation, and channel decoding 400. In FIG. 4, the channel estimation process may include the following. The receiver may determine the channel estimates of the DMRS symbols from their known locations in the received slots. An averaging window may be used to minimize the effects of noise. Multi-dimensional interpolation operations may be then used to estimate the missing values associated with one or more of the other resource elements (REs) (e.g., all the other REs) from the channel estimation grid. Noise power estimation may be performed to improve performance by comparison of direct and average channel estimates. Based on thechannel and noise estimates, the terminal then may design an equalizer (e.g., MMSE) followed by coherent demodulation and channel decoding.

[0096] When an AI / ML-based channel estimator for demodulation is trained (e.g., either online or offline) and then deployed for inference after training, its performance may degrade over time due to model drift. The deployed AI / ML model may be constantly receiving new data to make predictions upon. The new data received may be assumed to have similar probability distribution to the one used for training. However, this new data may have a different probability distribution than the one used for training. Hence, using the original AI / ML model with the new data distribution may cause a degradation in the model performance. In short, AI / ML model drift may be a situation where a model’s performance degrades over time, causing the model to start giving less accurate predictions. In such a situation, the performance of the AI / ML model may not be guaranteed and / or has some uncertainty in generalization.

[0097] A method for AI / ML model performance monitoring of channel estimation for demodulation may be discussed. When an AI / ML model is trained (e.g., either online or offline) for a certain receiver function, and then deployed for inference after training, it may be necessary that either the WTRU or the gNB performs the AI / ML model performance monitoring. Performance monitoring may enable the WTRU or the gNB deciding whether to continue using the existing model or to update / finetune the existing model with additional training. Although the proposed solution is applicable for any function at the receiver (e.g., channel state information (CSI) estimation, equalization, demapping, and the combination of them, etc.), for the purpose of illustration, the rest of the disclosure may focus on CSI estimation for demodulation. Considering the channel estimator for demodulation, this solution may focus on providing steps that allow monitoring the performance of AI / ML-based channel estimator. The summary of the solution may be listed below.

[0098] A WTRU may be configured and enabled to use an AI / ML model in the channel estimation for demodulation. The WTRU may apply preprocessing to the received DMRS (having been transmitted by the associated gNB) carrying REs after the fast fourier transform (FFT) of the receiver. The preprocessing may include division by (or multiplication by conjugate of) the known DMRS symbols to estimate the channel at theDMRS REs and / or other processing. The other processing may include, for example, a filtering option of the DMRS carrying REs. The input of the AI / ML model may be the preprocessed received DMRS. The output of the AI / ML model may be the effective channel estimate at each RE. The training of the AI / ML model may be performed either offline or online.

[0099] The WTRU may be configured to check whether the preprocessed received DMRS belong to the same probability distribution of the ones used for training the AI / ML model. The WTRU may be configured to apply an OOD detection algorithm to the preprocessed received DMRS REs prior to using the AI / ML model to check whether the input of the AI / ML-based channel estimator for demodulation is in-distribution or out-of- distribution. Examples of conventional OOD detection algorithms used in the literature may include, but not be limited to classification, probability and generative models, and / or reconstruction. In classification OOD detection algorithm, the original dataset of in-distribution examples of the preprocessed received DMRS REs may be used to build a one-class classifier that identifies if a new example belongs to the same dataset. In the OOD detection algorithm of probability and generative models, a probability model of the in-distribution preprocessed received DMRS REs may be built and new examples with low probability may be classified as out-of-distribution. In reconstruction OOD detection algorithm, the algorithm may learn an autoencoder that compresses the indistribution preprocessed received DMRS REs to a low dimensional representation and then reconstruct it from this representation. Out-of-distribution examples may be identified as those where the reconstruction quality is low.

[0100] The result of the OOD detection test may be either binary (e.g., either 0 or 1 ) or non-binary. Non-binary may include a floating-point number between [0, 1 ], In one example, the OOD detection test result may be binary. If the result of the OOD detection test is 0, for example, in-distribution, the WTRU may continue to use the AI / ML-based channel estimator to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the doppler estimate. If the result of the OOD detection test is 1 , for example, out-of-distribution, the WTRU may fall back to legacy / non-AI / ML CSI estimator to obtain the effective channel estimate at each RE and / or the effective noise variance and / or the doppler estimate. The WTRU mayattempt to determine if the 00D is due to a non-recom mended precoder. The WTRLI may be configured to report to the NW through the UCI that an OOD event occurred. The gNB may know that the OOD event is caused by a choice of a precoder that is not recommended by the WTRU. The gNB may inform the WTRLI that the OOD event is caused by a change of recommended precoder. In another example, the OOD detection test result may be a floating-point number between [0, 1] that approximates a probability of OOD. The WTRU may be configured with a threshold (or multiple thresholds) for comparison of the OOD result to decide which channel estimator to use. If the OOD probability is lower than the threshold, for example, in-distribution, the WTRU may continue to use the AI / ML-based channel estimator to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the doppler estimate. If the OOD probability is higher than the threshold, for example, out-of- distribution, the WTRU may fall back to legacy / non-AI / ML OS I estimator to obtain the effective channel estimate at each RE and / or the effective noise variance and / or the doppler estimate. The WTRU may attempt to determine if the OOD is due to a nonrecommended precoder. The WTRU may be configured to report to the network (NW) through the UCI that an OOD event occurred. The gNB may know that the OOD event is caused by a choice of a precoder that is not recommended by the WTRU. The gNB may inform the WTRU that the OOD event is caused by a change of recommended precoder.

[0101] In another solution, when channel state information reference signal (CSI-RS) allocations are transmitted, the WTRU may be configured to detect OOD via observation of the CSI-RS or the CSI derived from the CSI-RS. The WTRU may be configured to apply OOD detection for the channel estimated at the preprocessed received CSI-RS REs. The WTRU and / or the gNB may detect if the preprocessed received DMRS REs are in-distribution or out-of-distribution based on the OOD results of the preprocessed received CSI-RS REs. If the preprocessed received CSI-RS REs are in-distribution, the WTRU may assume the channel estimated at the preprocessed received DMRS REs are also in-distribution. The WTRU may continue to use the AI / ML model for channel estimation for demodulation to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the doppler estimateassociated to the preprocessed received DM RS REs. If the preprocessed received CSI- RS REs are out-of-distribution, the WTRU may assume the preprocessed received DMRS REs are also out-of-distribution. The WTRU may fall back to legacy / non-AI / ML channel estimator to estimate the effective channel at each RE and / or the effective noise variance and / or the doppler estimate associated to the preprocessed received DMRS REs. The WTRU may be configured to report to the NW through the UCI that an 00 D event occurred.

[0102] In another solution, the WTRU may be also configured to detect OOD via observation of other RSs in addition to DMRS and CSI-RS.

[0103] The WTRU may be configured to monitor the performance of the AI / ML model for channel estimation for demodulation.

[0104] The WTRU may be configured to receive the training data information from the gNB. For example, the training data information from the gNB may include the information of the preprocessed received DMRS REs used for training the AI / ML model. The training data information may include, but not be limited to, a set of preprocessed received DMRS REs (e.g., input of the AI / ML model) that were used initially in training the AI / ML model prior to deployment for online inference. The training data information may also include, but not be limited to statistics (e.g., cumulative density function (CDF), statistical mean, standard deviation, statistical moments, etc.) of the preprocessed received DMRS REs (e.g., input of the AI / ML model) that are used initially in training the AI / ML model. These statistics may be obtained by running statistical measurements on the preprocessed received DMRS REs dataset that is used in training the AI / ML model.

[0105] Whenever the WTRU receives DMRS to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the doppler estimate, the WTRU may be configured to apply the following steps. The WTRU may apply the preprocessing to the received DMRS REs. The WTRU may store the corresponding preprocessed received DMRS REs within a set with a preconfigured size (e.g., N) that contains the historical / past preprocessed received DMRS REs. The WTRU may run the OOD detection test to the preprocessed received DMRS REs and store the result within a set with a preconfigured size (e.g., N) that contains the OOD results of the historical / past preprocessed received DMRS REs.

[0106] The WTRU may use the set of preprocessed received DMRS REs, the set of corresponding OOD detection test results, and / or the training data information to run a model drift detection algorithm. The drift detection algorithm may consist of running a statistical test. The statistical test may include the Kolmogorov-Smirnov test, the Kullback-Leibler divergence, the Jensen-Shannon distance, the Page-Hinkley method, and the like.

[0107] In another solution, when CSI-RS allocations are transmitted concurrently with DMRS within the same resource grid, the WTRU may be configured to use channel estimation for sounding with an AI / ML model. The input of the AI / ML model may be the preprocessed received CSI-RS. The output of the AI / ML model may be the estimated full channel. The WTRU may be configured to monitor the performance of the AI / ML model for CSI estimation for sounding. The WTRU and / or the gNB may detect if the AI / ML model for CSI estimation for demodulation is drifting or not based on the drift detection results of the AI / ML model for channel estimation for sounding. If the AI / ML model for CSI estimation for sounding is not drifting, the WTRU may assume that the AI / ML model for CSI estimation for demodulation is also not drifting. If the AI / ML model for CSI estimation for sounding is drifting, the WTRU may assume that the AI / ML model for CSI estimation for demodulation is also drifting.

[0108] Either the WTRU or the gNB may proceed with the following predefined actions based on the AI / ML model drift detection outcomes enumerated below. If the AI / ML model has not drifted and is not drifting, for example, the AI / ML model drift is below the first threshold, the WTRU may be configured to keep using the same AI / ML model. If the AI / ML model is drifting but the total drift is above the first threshold and below the second threshold, the WTRU may be configured to request the gNB to initiate updating / retraining the AI / ML model. If the AI / ML model has drifted, for example, the total drift is above the second threshold, the WTRU may be configured to request the gNB to initiate updating / retraining the AI / ML model, while also using the fallback (e.g., non-AIML) algorithm for channel estimation for demodulation.

[0109] The WTRU may be configured to report the results of the model drift detection algorithm to the gNB. The drift detection results may be reported to the gNB periodically or aperiodically in the UCI that can be carried either by physical uplink control channel(PUCCH) or physical uplink shared channel (PUSCH). For periodic drift detection reporting, the WTRU may be configured by the NW via higher layer signaling. The higher layer signaling may include RRC signaling, system information block (SIB), and the like. For aperiodic drift detection reporting, the WTRU may be configured by the NW via higher layer signaling (e.g., RRC signaling, SIB, and the like) and may be triggered by DCI or medium access control element (MAC CE). If the AI / ML model has not drifted and is not drifting, for example, the AI / ML model drift is below the first threshold: the gNB may be configured to inform the WTRU to keep using the same AI / ML model. If the AI / ML model is drifting but the total drift is above the first threshold and below the second threshold, the gNB may be configured to initiate updating / retraining the AI / ML model and may inform the WTRU to keep using the same AI / ML model. If the AI / ML model has drifted, for example, the total drift is above the second threshold, the gNB may be configured to initiate updating / retraining the AI / ML model and inform the WTRU to use the fallback (e.g., non-AIML) algorithm for channel estimation for demodulation.

[0110] When an AI / ML model is trained (e.g., either online or offline) for a certain receiver function, and then deployed for inference after training, it may be necessary that either the WTRU or the gNB performs the AI / ML model performance monitoring. Performance monitoring may enable the WTRU or the gNB deciding whether to continue using the existing model or to update / finetune the existing model with additional training. Although the proposed solution is applicable for any function at the receiver (e.g., CSI estimation, equalization, demapping, and the combination of them, etc.), for the purpose of illustration, the rest of the disclosure may focus on CSI estimation for demodulation (e.g., focus only on CSI estimation for demodulation) as an example. Considering the channel estimator for demodulation, this solution may focus on providing steps that allow monitoring the performance of AI / ML-based channel estimator.

[0111] The WTRU may perform the downlink channel estimation for demodulation via an AI / ML model. The stages of the AI / ML process may be detailed as follows. The application of AI / ML model is the channel estimation for demodulation. The input data may include the received DMRS REs. The preprocessing may include, but not be limited to extraction of the received DMRS REs from the received signals using theconfigured DMRS locations (e.g., allocations for the downlink), division of the received DMRS REs by known DMRS pilots, multiplication of the received DMRS signals by the conjugate of the known DMRS pilots, applying a 2D filtering to the DMRS carrying REs, reshaping the input data, and / or concatenation of the real and imaginary parts to obtain the real-valued input to the AI / ML model.

[0112] In the literature, various AI / ML models may have been proposed for the channel estimation (e.g., ReEsNet, ChannelNet, DDAE, MTRE). The results in the literature may show that the ReEsNet model achieves higher channel estimation accuracy with low AI / ML model complexity (e.g., less trainable parameters) compared to its benchmarks. FIG. 5 illustrates an example of a ReEsNet model 500. The ReEsNet model architecture may be constructed via convolutional layers, rectified linear unit (ReLU) activation layers, and residual connections.

[0113] The postprocessing may include, but not be limited to conversion of the real- valued AI / ML model output to the complex-valued output data, and / or reshaping the AI / ML model output.

[0114] Effective channel may estimate at each RE associated to the preprocessed received DMRS REs at the AI / ML model input.

[0115] The training may be performed either offline or online in a supervised manner. The features may be the preprocessed received DMRS REs. The labels may be the error-free ground-truth effective channel estimates at each RE.

[0116] When a WTRU is configured to perform OOD detection test to the preprocessed received DMRS REs that are input to the AI / ML-based channel estimation for demodulation, key parameters may be configured by the NW via higher layer signaling. The higher layer signaling may include RRC signaling, SIB, and the like. The configuration may include, but not be limited to, one or more information elements and parameters, The one or more information elements and parameters may include the OOD detection algorithm, the reference distribution of the preprocessed received DMRS used in training the AI / ML model, and / or the OOD probability threshold.

[0117] When a WTRU is configured to perform drift detection of the AI / ML model used for the channel estimation for demodulation, key parameters may be configured by the NW via higher layer signaling (e.g., RRC signaling, SIB, etc.) The configuration mayinclude, but not be limited to, the following information elements and parameters: the AI / ML model drift algorithm, the drift detection threshold, a set of preprocessed received DMRS REs used in training the AI / ML model, the statistics (e.g., CDF, statistical mean, standard deviation, statistical moments, etc.) of the preprocessed received DMRS REs that were used in training the AI / ML model, and / or the size (N) or received DMRS set used for drift detection by the WTRU.

[0118] FIG. 6 illustrates the procedures for the WTRU to perform OOD detection test in channel estimation for demodulation 600.

[0119] At 602, the WTRU may apply preprocessing to the received DMRS. The preprocessing may include, but not be limited to extraction of the received DMRS REs from the received signals using the configured DMRS locations (e.g., allocations for the downlink), division of the received DMRS REs by known DMRS pilots, multiplication of the received DMRS signals by the conjugate of the known DMRS pilots, applying a 2D filtering to the DMRS carrying REs.

[0120] At 604, the WTRU may be configured to apply an OOD detection algorithm to the preprocessed received DMRS REs prior to using the AI / ML model to check whether the input of the AI / ML-based channel estimator for demodulation is in-distribution or out-of- distribution. FIG. 7 illustrates a DMRS OOD detection mechanism in the WTRU 700.Examples of conventional OOD detection algorithms used in the literature may include, but not be limited to classification, probability and generative models, and / or reconstruction. In classification OOD detection algorithm, the original dataset of indistribution examples of the preprocessed received DMRS REs may be used to build a one-class classifier that identifies if a new example belongs to the same dataset. In probability and generative models, a probability model of the in-distribution preprocessed received DMRS REs may be built and new examples with low probability may be classified as out-of-distribution. In reconstruction OOD detection algorithm, the algorithm may learn an autoencoder that compresses the in-distribution preprocessed received DMRS REs to a low dimensional representation and then may reconstruct it from this representation. For example, out-of-distribution examples may be identified as those where the reconstruction quality is low.

[0121] At 606, the WTRU may determine whether the OOD detection test produces binary output. The result of the OOD detection test may be either binary (e.g., either 0 or 1 ) or non-binary, such as a floating-point number between [0, 1 ],

[0122] If the OOD detection test result is binary, the WTRU may determine whether the output is 0 or 1 at 608. If the result of the OOD detection test is 0, for example, indistribution, the WTRU may continue to use the AI / ML-based channel estimator to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the Doppler estimate at 612. If the result of the OOD detection test is 1 , for example, out-of-distribution, the WTRU may fall back to legacy / non-AI / ML CSI estimator to obtain the effective channel estimate at each RE and / or the effective noise variance and / or the Doppler estimate at 614. The WTRU may attempt to determine if the OOD is due to a non-recom mended precoder. The WTRU may be configured to report to the NW through the UCI that an OOD event occurred. The gNB may know that the OOD event is caused by a choice of a precoder that is not recommended by the WTRU. The gNB may inform the WTRU that the OOD event is caused by a change of recommended precoder.

[0123] If the OOD detection test result is a floating-point number between [0, 1] that approximates a probability of OOD, the WTRU may determine whether the floating-point number output is above a threshold at 610. The WTRU may be configured with a threshold, or multiple thresholds, for comparison of the OOD result to decide which channel estimator to use. If the OOD probability is lower than the threshold, for example, in-distribution, the WTRU may continue to use the AI / ML-based channel estimator to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the Doppler estimate at 612. If the OOD probability is higher than the threshold, for example, out-of-distribution, the WTRU may fall back to legacy / non-AI / ML CSI estimator to obtain the effective channel estimate at each RE and / or the effective noise variance and / or the Doppler estimate at 61 . The WTRU may attempt to determine if the OOD is due to a non-recom mended precoder. The WTRU may be configured to report to the NW through the UCI that an OOD event occurred. The gNB may know that the OOD event is caused by a choice of a precoder that is notrecommended by the WTRU. The gNB may inform the WTRLI that the OOD event is caused by a change of recommended precoder.

[0124] In another solution, when CSI-RS allocations are transmitted, the WTRU may be configured to detect OOD via observation of the CSI-RS or the CSI derived from the CSI-RS. The WTRU may be configured to apply OOD detection for the channel estimated at the preprocessed received CSI-RS REs. The WTRU or the gNB may detect if the preprocessed received DMRS REs are in-distribution or out-of-distribution based on the OOD results of the preprocessed received CSI-RS Res. If the preprocessed received CSI-RS REs are in-distribution, the WTRU may assume the channel estimated at the preprocessed received DMRS REs are also in-distribution.The WTRU may use the AI / ML model for channel estimation for demodulation to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the Doppler estimate associated to the preprocessed received DMRS REs at 612. If the preprocessed received CSI-RS REs are out-of-distribution, the WTRU may assume the preprocessed received DMRS REs are also out-of-distribution. The WTRU may fall back to legacy / non-AI / ML channel estimator to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the Doppler estimate associated to the preprocessed received DMRS REs at 614. The WTRU may be configured to report to the NW through the UCI that an OOD event occurred.

[0125] In another solution, the WTRU may be also configured to detect OOD via observation of other RSs in addition to DMRS and CSI-RS.

[0126] FIGs. 8A-8C illustrate the steps and procedures for the WTRU to perform AI / ML model drift detection in channel estimation for demodulation.

[0127] The WTRU may be configured to monitor the performance of the AI / ML model for channel estimation for demodulation.

[0128] At 802, the WTRU may be configured to receive the training-data information from the gNB, for example, the information of the preprocessed received DMRS REs used for training the AI / ML model. The training-data information may include, but not be limited to a set of preprocessed received DMRS REs (input of the AI / ML model) that were used initially in training the AI / ML model prior to deployment for online inference, and / or the statistics (e.g., CDF, statistical mean, standard deviation, statisticalmoments, etc.) of the preprocessed received DMRS REs (input of the AI / ML model) that are used initially in training the AI / ML model. These statistics may be obtained by running statistical measurements on the preprocessed received DMRS REs dataset that is used in training the AI / ML model.

[0129] Whenever the WTRU receives DMRS to predict the effective channel estimate at each RE and / or the effective noise variance estimate and / or the doppler estimate, the WTRU may be configured to apply AI / ML model drift detection mechanism. FIG. 9 illustrates the AI / ML model drift detection mechanism in the WTRU 900. The AI / ML model drift detection mechanism may include the following steps. The WTRU may apply the preprocessing to the received DMRS REs. The preprocessing may include, but not be limited to extraction of the received DMRS REs from the received signals using the configured DMRS locations (e.g., allocations for the downlink), division of the received DMRS REs by known DMRS pilots, multiplication of the received DMRS signals by the conjugate of the known DMRS pilots, applying a 2D filtering to the DMRS carrying REs. The WTRU may store the corresponding preprocessed received DMRS REs within a set with a preconfigured size (N) that contains the historical / past preprocessed received DMRS REs at 804. The WTRU may run the OOD detection test to the preprocessed received DMRS REs and store the result within a set with a preconfigured size (N) that contains the OOD results of the historical / past preprocessed received DMRS REs at 806.

[0130] The WTRU may use the set of the preprocessed received DMRS REs, the set of corresponding OOD detection test results, and the training data information to run a model drift detection algorithm at 808. The drift detection algorithm may consist of running a statistical test, such as, the Kolmogorov-Smirnov test, the Kullback-Leibler divergence, the Jensen-Shannon distance, the Page-Hinkley method, and the like.

[0131] In another solution, CSI-RS allocations may be transmitted concurrently with DMRS within the same resource grid. The WTRU may be configured to use channel estimation for sounding with an AI / ML model. The input of the AI / ML model may be the preprocessed received CSI-RS. The output of the AI / ML model may be the estimated full channel. The WTRU may be configured to monitor the performance of the AI / ML model for CSI. The WTRU or the gNB may detect if the AI / ML model for CSI estimationfor demodulation is drifting or not based on the drift detection results of the AI / ML model for channel estimation for sounding: If the AI / ML model for CSI estimation for sounding is not drifting, the WTRU may assume that the AI / ML model for CSI estimation for demodulation is also not drifting. If the AI / ML model for CSI estimation for sounding is drifting, the WTRU may assume that the AI / ML model for CSI estimation for demodulation is also drifting.

[0132] The WTRU may determine whether the WTRU is configured to send test results to the gNB at 810. Either the WTRU or the gNB may proceed with the following predefined actions based on the AI / ML model drift detection outcomes enumerated as follows.

[0133] If the WTRU is configured to proceed, the WTRU may determine whether the AI / ML model is drifting at 812. The WTRU may use two thresholds that are configured by the gNB. If the AI / ML model is drifting, the WTRU may then determine whether the drift is above a first threshold at 816. If the AI / ML model has not drifted and is not drifting, for example, the AI / ML model drift is below the first threshold, the WTRU may keep using the same AI / ML model at 820. If the WTRU determines that the drift is above the first threshold, the WTRU may then determine whether the drift is above a second threshold at 832. If the AI / ML model is drifting but the total drift is above the first threshold and below the second threshold, the WTRU may keep using the same AI / ML model at 820 and may request the gNB to initiate updating / retraining the AI / ML model at 822. If the AI / ML model has drifted, for example, the total drift is above the first threshold and the second threshold, the WTRU may request the gNB to initiate updating / retraining the AI / ML model at 822, while also using the fallback (e.g., non- AIML) algorithm for channel estimation for demodulation at 824.

[0134] If the gNB is configured to proceed, the WTRU may determine whether the AI / ML model is drifting at 814. If the AI / ML model is drifting, the WTRU may then determine whether the drift is above a first threshold at 818. The WTRU may be configured to report the results of the model drift detection algorithm to the gNB. The drift detection results may be reported to the gNB periodically or aperiodically in the UCI that can be carried either by PUCCH or PUSCH. For periodic drift detection reporting, the WTRU may be configured by the NW via higher layer signaling, such as RRC signaling, SIB,and the like. For aperiodic drift detection reporting, the WTRU may be configured by the NW via higher layer signaling, such as RRC signaling, SIB, and the like, and may be triggered by DCI or MAC CE. If the AI / ML model has not drifted and is not drifting, for example, the AI / ML model drift is below the first threshold, the gNB may inform the WTRU to keep using the same AI / ML model at 826. If the AI / ML model is drifting and the drift is above the first threshold, the WTRU may then determine whether the drift is above a second threshold at 834. If the AI / ML model is drifting but the total drift is above the first threshold and below the second threshold, the gNB may initiate updating / retraining the AI / ML model at 828 and may inform the WTRU to keep using the same AI / ML model at 826. If the AI / ML model has drifted, for example, the total drift is above the first threshold and the second threshold, the gNB may initiate updating / retraining the AI / ML model 828 and inform the WTRU to use the fallback (e.g., non-AI / ML) algorithm for channel estimation for demodulation at 830.

Claims

CLAIMS:1 . A wireless transmit / receive unit (WTRU) comprising: a processor configured to: receive, from a network device, training data information related to an artificial intelligence / machine learning model, wherein the training data information comprises information related to demodulation reference signal (DMRS) resource elements (REs) that were used to train the model; determine that the model has drifted based on (i) a set of preprocessed received DMRS REs, (ii) results of an out-of-distribution (OOD) detection test performed on the set of preprocessed received DMRS REs, and (iii) the training data information; and send an indication that the artificial intelligence / machine learning model has drifted to the network device.

2. The WTRU of claim 1 , wherein the indication that the model has drifted comprises results of a model drift detection test performed by the WTRU.

3. The WTRU of claim 1 , wherein the results of the OOD detection test comprise a binary result of the OOD detection test.

4. The WTRU of claim 1 , wherein the training data information comprises statistics associated with the preprocessed received DMRS REs that were used to train the model.

5. The WTRU of claim 1 , wherein the processor is configured to: receive a set of DMRS REs; and preprocess the set of received DMRS REs to generate the set of preprocessed received DMRS REs.

6. The WTRU of claim 1 , wherein the processor is configured to:determine that the model has drifted further based on a comparison of an output of a drift detection test to a threshold, wherein the threshold is received from the network device.

7. The WTRU of claim 1 , wherein the processor is configured to: send a request to the network device to receive updated information for the model.

8. The WTRU of claim 1 , wherein the processor is configured to: send a request the network device to receive updated information for the model; and use a fallback algorithm for channel estimation for demodulation until the WTRU receives the updated information for the model.

9. The WTRU of claim 1 , wherein the processor is configured to: receive updated information to update the model, wherein the indication that the model has drifted comprises results of the model drift detection algorithm.

10. The WTRU of claim 1 , wherein the processor is configured to: determine that the model has drifted using a drift detection test, wherein the drift detection test comprises a Kolmogorov-Smirnov test, a Kullback-Leibler divergence test, a Jensen-Shannon distance test, or a Page-Hinkley test.

11. A method implemented by a wireless transmit / receive unit (WTRU), the method comprising: receiving, from a network device, training data information related to an artificial intelligence / machine learning model, wherein the training data information comprises information related to demodulation reference signal (DMRS) resource elements (REs) that were used to train the model; determining that the model has drifted based on (i) a set of preprocessed received DMRS REs, (ii) results of an out-of-distribution (OOD) detection test performedon the set of preprocessed received DMRS REs, and (iii) the training data information; and sending an indication that the artificial intelligence / machine learning model has drifted to the network device.

12. The method of claim 11 , wherein the indication that the model has drifted comprises results of a model drift detection test performed by the WTRLI.

13. The method of claim 11 , wherein the results of the OOD detection test comprise a binary result of the OOD detection test.

14. The method of claim 11 , wherein the training data information comprises statistics associated with the preprocessed received DMRS REs that were used to train the model.

15. The method of claim 11 , further comprising: receiving a set of DMRS REs and preprocessing the set of received DMRS REs to generate the set of preprocessed received DMRS REs.

16. The method of claim 11 , further comprising: determining that the model has drifted further based on a comparison of an output of a drift detection test to a threshold, wherein the threshold is received from the network device.

17. The method of claim 11 , further comprising: sending a request to the network device to receive updated information for the model.

18. The method of claim 11 , further comprising: sending a request the network device to receive updated information for the model; andusing a fallback algorithm for channel estimation for demodulation until the WTRLI receives the updated information for the model.

19. The method of claim 11 , further comprising: receiving updated information to update the model, wherein the indication that the model has drifted comprises results of the model drift detection algorithm.

20. The method of claim 11 , further comprising: determining that the model has drifted using a drift detection test, wherein the drift detection test comprises a Kolmogorov-Smirnov test, a Kullback-Leibler divergence test, a Jensen-Shannon distance test, or a Page-Hinkley test.

Images