Monitoring to which performance prediction model of ai / ML-based fallback operation is applied
Patent Information
- Application Number
- PCT/KR2026/004426
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-03-24
- Filing Date
- 2026-03-19
- Publication Date
- 2026-10-01
Smart Images
Figure KR2026004426_01102026_PF_FP_ABST
Abstract
Description
Monitoring applying an AI / ML-based alternative behavior performance prediction model
[0001] The present disclosure relates to monitoring using an AI (Artificial Intelligence) / ML (Machine Learning) based performance prediction model for alternative actions.
[0002] 3GPP (3rd Generation Partnership Project) New Radio (NR) targets a single technical framework that addresses all deployment, use, and requirements, including enhanced Mobile Broadband (eMBB), massive Machine Type Communications (mmTC), and Ultra-Reliable and Low Latency Communications (URLLC). Additionally, NR must be able to utilize any spectrum band up to at least 100 GHz that can be used for wireless communication in the distant future. NR must be inherently forward compatible.
[0003] 6G is the successor to 5G cellular technology. 6G networks can utilize higher frequencies than 5G networks and will provide significantly higher capacity and much lower latency. The 6G technology market is expected to drive massive improvements in imaging, presence technology, and location awareness. Working in conjunction with Artificial Intelligence (AI), 6G computing infrastructure will be able to identify the best places for computing to occur. This includes decisions regarding data storage, processing, and sharing.
[0004] In 5G and 6G mobile communication systems, research is actively underway to apply AI (Artificial Intelligence) / ML (Machine Learning) techniques to various fields, such as channel state prediction, beam management, positioning accuracy improvement, and network optimization. Accordingly, standardization organizations including 3GPP are discussing frameworks for the Life Cycle Management (LCM) of AI / ML models, and a series of procedures such as model training, deployment, monitoring, and updating are being standardized.
[0005] However, the conventional technology has the following problems.
[0006] First, there are no established criteria for determining subsequent fallback operations when the performance of an AI / ML model degrades. Generally, AI / ML models can experience performance degradation over time due to discrepancies between the distribution of training data and the actual operating environment, namely data drift or concept drift. Even if the management function detects this degradation, if the criteria for selecting a fallback operation are not clearly defined, unnecessary LCM operations (e.g., model retraining, rollback, or replacement procedures) may be frequently triggered, leading to increased model management overhead. Furthermore, excessive LCM operations result in the waste of system resources and can negatively impact the provision of stable services.
[0007] In addition, depending on the mobility of the terminal and rapid changes in the channel environment, the optimal alternative action applicable on the terminal side or the network side may vary depending on the situation. For example, in an environment where the terminal moves at high speed, channel changes occur rapidly, so a specific alternative action may not be effective, whereas conversely, in a static environment, a different alternative action may be more suitable. However, in conventional technology, there is no mechanism to select an alternative action by reflecting such dynamic environmental changes, and there is a problem in that the adaptability and efficiency of the system are reduced because the alternative action is determined according to uniform criteria.
[0008] To address the aforementioned problem, a method may be required to improve the generalization performance of AI / ML models by introducing an evaluation model that predicts the performance of alternative actions in advance and optimizing subsequent action decisions based on this.
[0009] In one embodiment, a method performed by User Equipment (UE) is provided. The method includes the step of determining performance degradation of an operation, and the step of reporting to a network information related to one or more alternative operations applicable to said operation. Performance prediction values of said one or more alternative operations are derived based on an evaluation model. The method includes the step of receiving from the network configuration information related to an optimal alternative operation selected based on the performance prediction values of said one or more alternative operations, and the step of executing said optimal alternative operation based on said configuration information.
[0010] In one embodiment, User Equipment (UE) is provided. The UE includes at least one transceiver, at least one processor, and at least one memory that stores instructions that may be operablely connected to the at least one processor and cause the UE to perform an operation based on execution by the at least one processor. The operation includes the steps of determining performance degradation of the operation and reporting information related to one or more alternative operations applicable to the operation to a network. Performance prediction values of the one or more alternative operations are derived based on an evaluation model. The method includes the steps of receiving configuration information related to an optimal alternative operation selected based on performance prediction values of the one or more alternative operations from the network, and executing the optimal alternative operation based on the configuration information.
[0011] In one embodiment, a device is provided. The device includes at least one processor integrated with a User Equipment (UE), and at least one memory storing processor-executable instructions configured to enable the at least one processor to perform an operation. The operation includes the step of determining performance degradation of the operation, and the step of generating information related to one or more alternative operations applicable to the operation. Performance prediction values of the one or more alternative operations are derived based on an evaluation model. The method includes the step of obtaining configuration information related to an optimal alternative operation selected based on performance prediction values of the one or more alternative operations, and the step of executing the optimal alternative operation based on the configuration information.
[0012] In one embodiment, a non-transitory computer-readable medium (CRM) is provided for storing instructions that perform an operation based on execution by at least one processor. The operation includes the steps of determining performance degradation of the operation and reporting information related to one or more alternative operations applicable to the operation to a network. Performance prediction values of the one or more alternative operations are derived based on an evaluation model. The method includes the steps of receiving configuration information related to an optimal alternative operation selected based on performance prediction values of the one or more alternative operations from the network, and executing the optimal alternative operation based on the configuration information.
[0013] In one embodiment, a method performed by a base station is provided. The method includes the step of receiving from a UE information related to one or more alternative actions applicable to a degraded action. Performance prediction values of the one or more alternative actions are derived based on an evaluation model. The method includes the step of transmitting to the UE configuration information related to an optimal alternative action selected based on the performance prediction values of the one or more alternative actions. Based on the configuration information, the optimal alternative action is executed.
[0014] In one embodiment, a base station is provided. The base station includes at least one transceiver, at least one processor, and at least one memory that may be operably connected to the at least one processor and stores instructions that cause the UE to perform an operation based on execution by the at least one processor. The operation includes receiving from the UE information related to one or more alternative operations applicable to the operation with degraded performance. Performance prediction values of the one or more alternative operations are derived based on an evaluation model. The method includes transmitting to the UE configuration information related to an optimal alternative operation selected based on the performance prediction values of the one or more alternative operations. The optimal alternative operation is executed based on the configuration information.
[0015] The present disclosure may have various effects.
[0016] For example, if performance degradation of the main model is detected, an intelligent action decision system differentiated from conventional monitoring methods can be provided by comparing and / or evaluating the predictive performance of all conceivable alternative actions in real time.
[0017] For example, when determining an alternative action, one or more types and / or categories of alternative actions that are operable or preferred by the network side or the terminal may be considered. Through this, an optimal alternative action that matches the capabilities and preferences of the network environment and / or the terminal can be selected, and more flexible and adaptive model management is possible compared to conventional methods that rely on uniform criteria.
[0018] For example, by post-configuring multiple legacy operations in terms of fallback and comparing their respective predictive performance, it is possible to support a transition to a legacy operation optimized for the current operating environment. Accordingly, unnecessary periods of performance degradation can be minimized, and service continuity can be improved.
[0019] For example, in terms of model switching, it is possible to compare and / or evaluate the performance of multiple candidate AI / ML models in real time. This supports dynamic switching to the model best suited to the current channel environment and terminal conditions, and can improve overall system performance and robustness compared to conventional methods that rely on a single model.
[0020] For example, even if the performance of the optimized alternative action itself degrades, it is possible to search for and determine the next-best alternative action by recursively applying the monitoring process. Based on this recursive monitoring structure, the system can continuously maintain an optimal operating state even in situations of multi-stage performance degradation.
[0021] The effects obtainable through the specific examples of the present disclosure are not limited to those listed above. For example, there may be various technical effects that a person having ordinary skill in the related art can understand or derive from the present disclosure. Accordingly, the specific effects of the present disclosure are not limited to those explicitly described in the present disclosure, but may include various effects that can be understood or derived from the technical features of the present disclosure.
[0022] FIG. 1 shows an example of a communication system to which an implementation of the present disclosure is applied.
[0023] FIG. 2 shows an example of a wireless device to which an implementation of the present disclosure is applied.
[0024] FIG. 3 shows an example of a UE to which an implementation of the present disclosure is applied.
[0025] FIG. 4 shows an example of a frame structure in a 3GPP-based wireless communication system to which an implementation of the present disclosure is applied.
[0026] FIG. 5 shows an example of DL RS transmission of a base station to which an implementation of the present disclosure is applied.
[0027] FIG. 6 shows an example of a DL BM procedure using an SSB to which an implementation of the present disclosure is applied.
[0028] FIG. 7 shows an example of a DL BM procedure using CSI-RS to which an implementation of the present disclosure is applied.
[0029] FIG. 8 shows an example of a receiving beam determination procedure of a UE to which an implementation of the present disclosure is applied.
[0030] FIG. 9 shows an example of a transmission beam determination procedure of a base station to which an implementation of the present disclosure is applied.
[0031] FIG. 10 shows an example of a UL BM procedure using an SRS to which an implementation of the present disclosure is applied.
[0032] FIG. 11 shows an example of a transmit beam determination procedure of a UE to which an implementation of the present disclosure is applied.
[0033] FIG. 12 shows an example of a functional framework of AI / ML to which an implementation of the present disclosure is applied.
[0034] FIG. 13 illustrates an example of a method performed by a UE applicable to an implementation of the present disclosure.
[0035] FIG. 14 illustrates an example of a method performed by a base station applicable to an implementation of the present disclosure.
[0036] FIG. 15 illustrates an example of a procedure to which an implementation of the present disclosure is applied.
[0037] FIG. 16 illustrates another example of a procedure to which an implementation of the present disclosure is applied.
[0038] The following techniques, devices, and systems may be applied to various wireless multiple access systems. Examples of multiple access systems include Code Division Multiple Access (CDMA) systems, Frequency Division Multiple Access (FDMA) systems, Time Division Multiple Access (TDMA) systems, Orthogonal Frequency Division Multiple Access (OFDMA) systems, Single Carrier Frequency Division Multiple Access (SC-FDMA) systems, and Multi-Carrier Frequency Division Multiple Access (MC-FDMA) systems. CDMA may be implemented through wireless technologies such as Universal Terrestrial Radio Access (UTRA) or CDMA2000. TDMA may be implemented through wireless technologies such as Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), or Enhanced Data Rates for GSM Evolution (EDGE). OFDMA can be implemented through wireless technologies such as IEEE (Institute of Electrical and Electronics Engineers) 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, or E-UTRA (Evolved UTRA). UTRA is part of UMTS (Universal Mobile Telecommunications System). 3GPP (3rd Generation Partnership Project) LTE (Long-Term Evolution) is part of E-UMTS (Evolved UMTS) using E-UTRA.3GPP LTE uses OFDMA in the downlink (DL) and SC-FDMA in the uplink (UL). Evolutions of 3GPP LTE include LTE-A (Advanced), LTE-A Pro, and / or 5G NR (New Radio).
[0039] For convenience of explanation, the implementation of the present disclosure is described primarily in relation to 3GPP-based wireless communication systems. However, the technical characteristics of the present disclosure are not limited thereto. For example, the following detailed description is provided based on a mobile communication system corresponding to a 3GPP-based wireless communication system, but aspects of the present disclosure that are not limited to 3GPP-based wireless communication systems may be applied to other mobile communication systems.
[0040] For terms and technologies used in this disclosure that are not specifically described, reference may be made to wireless communication standard documents published prior to this disclosure.
[0041] In the present disclosure, "A or B" may mean "only A," "only B," or "both A and B." Alternatively, in the present disclosure, "A or B" may be interpreted as "A and / or B." For example, in the present disclosure, "A, B or C" may mean "only A," "only B," "only C," or "any combination of A, B and C."
[0042] A slash ( / ) or a comma used in the present disclosure may mean "and / or." For example, "A / B" may mean "A and / or B." Accordingly, "A / B" may mean "only A," "only B," or "both A and B." For example, "A, B, C" may mean "A, B or C."
[0043] In the present disclosure, "at least one of A and B" may mean "only A," "only B," or "both A and B." Additionally, in the present disclosure, the expressions "at least one of A or B" or "at least one of A and / or B" may be interpreted as synonymous with "at least one of A and B."
[0044] Additionally, in the present disclosure, "at least one of A, B and C" may mean "only A," "only B," "only C," or "any combination of A, B and C." Furthermore, "at least one of A, B or C" or "at least one of A, B and / or C" may mean "at least one of A, B and C."
[0045] Additionally, parentheses used in the present disclosure may mean "for example." Specifically, when indicated as "control information (PDCCH)," "PDCCH" may be proposed as an example of "control information." In other words, the "control information" of the present disclosure is not limited to "PDCCH," and "PDCCH" may be proposed as an example of "control information." Furthermore, even when indicated as "control information (i.e., PDCCH)," "PDCCH" may be proposed as an example of "control information."
[0046] Technical features described individually within one drawing in this disclosure may be implemented individually or simultaneously.
[0047] Although not limited thereto, the various descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure may be applied to various fields where wireless communication and / or connectivity between devices (e.g., 5G) is required.
[0048] The present disclosure will be described in more detail below with reference to the drawings. In the following drawings and / or description, the same reference numerals may refer to the same or corresponding hardware blocks, software blocks, and / or function blocks unless otherwise indicated.
[0049] The present disclosure will describe embodiments based on the structure and procedures, messages, etc. of a 5G mobile communication system. However, this is merely an example, and the embodiments of the present disclosure are not limited thereto. For example, the embodiments of the present disclosure can be extended to apply to an evolved form of a 6G mobile communication system. For example, the 5G-based messages described in the embodiments of the present disclosure may be defined as other existing messages, new messages, or parameters.
[0050] FIG. 1 shows an example of a communication system to which an implementation of the present disclosure is applied.
[0051] The 5G usage scenario shown in FIG. 1 is merely an example, and the technical features of the present disclosure may be applied to other 5G usage scenarios not shown in FIG. 1.
[0052] The three main requirement categories for 5G are (1) enhanced Mobile BroadBand (eMBB) category, (2) massive Machine Type Communication (mMTC) category, and (3) Ultra-Reliable and Low Latency Communications (URLLC) category.
[0053] Referring to FIG. 1, a communication system (1) includes wireless devices (100a to 100f), a base station (BS; 200), and a network (300). FIG. 1 illustrates a 5G network as an example of the network of the communication system (1), but the implementation of the present disclosure is not limited to a 5G system and can be applied to future communication systems beyond a 5G system.
[0054] The base station (200) and the network (300) can be implemented as wireless devices, and a specific wireless device can operate as a base station / network node in relation to another wireless device.
[0055] Wireless devices (100a to 100f) represent devices that perform communication using Radio Access Technology (RAT) (e.g., 5G NR or LTE) and may also be referred to as communication / wireless / 5G devices. Wireless devices (100a to 100f) may include, but are not limited to, robots (100a), vehicles (100b-1 and 100b-2), eXtended Reality (XR) devices (100c), portable devices (100d), home appliances (100e), Internet-Of-Things (IoT) devices (100f), and Artificial Intelligence (AI) devices / servers (400). For example, vehicles may include vehicles with wireless communication capabilities, autonomous vehicles, and vehicles capable of performing communication between vehicles. Vehicles may include unmanned aerial vehicles (UAVs) (e.g., drones). XR devices may include AR (Augmented Reality) / VR (Virtual Reality) / MR (Mixed Reality) devices and may be implemented in the form of HMDs (Head-Mounted Devices) and HUDs (Head-Up Displays) mounted on vehicles, televisions, smartphones, computers, wearable devices, home appliances, digital signs, vehicles, robots, etc. Portable devices may include smartphones, smart pads, wearable devices (e.g., smartwatches or smart glasses), and computers (e.g., laptops). Home appliances may include TVs, refrigerators, and washing machines. IoT devices may include sensors and smart meters.
[0056] In the present disclosure, wireless devices (100a to 100f) may be referred to as User Equipment (UE). The UE may include, for example, a mobile phone, a smartphone, a laptop computer, a digital broadcasting terminal, a PDA (Personal Digital Assistant), a PMP (Portable Multimedia Player), a navigation system, a slate PC, a tablet PC, an ultrabook, a vehicle, a vehicle with autonomous driving capabilities, a connected car, a UAV, an AI module, a robot, an AR device, a VR device, an MR device, a hologram device, a public safety device, an MTC device, an IoT device, a medical device, a fintech device (or financial device), a security device, a weather / environment device, a 5G service-related device, or a device related to the Fourth Industrial Revolution.
[0057] Wireless devices (100a to 100f) can be connected to a network (300) through a base station (200). AI technology may be applied to the wireless devices (100a to 100f), and the wireless devices (100a to 100f) can be connected to an AI server (400) through the network (300). The network (300) can be configured using a 3G network, a 4G (e.g., LTE) network, a 5G (e.g., NR) network, and a network after 5G. The wireless devices (100a to 100f) may communicate with each other through the base station (200) / network (300), but they may also communicate directly (e.g., sidelink communication) without going through the base station (200) / network (300). For example, vehicles (100b-1, 100b-2) can communicate directly (e.g., V2V (Vehicle-to-Vehicle) / V2X (Vehicle-to-everything) communication). Also, IoT devices (e.g., sensors) can communicate directly with other IoT devices (e.g., sensors) or other wireless devices (100a to 100f).
[0058] Wireless communication / connections (150a, 150b, 150c) can be established between wireless devices (100a to 100f) and / or between wireless devices (100a to 100f) and base station (200) and / or between base station (200). Here, the wireless communication / connections can be established through various RATs (e.g., 5G NR), such as uplink / downlink communication (150a), sidelink communication (150b) (or D2D (Device-To-Device) communication), and communication between base stations (150c) (e.g., relay, IAB (Integrated Access and Backhaul)). Through the wireless communication / connections (150a, 150b, 150c), wireless devices (100a to 100f) and base station (200) can transmit / receive wireless signals to / from each other. For example, wireless communication / connection (150a, 150b, 150c) may transmit / receive signals through various physical channels. To this end, based on various proposals of the present disclosure, at least some of the following may be performed: a process for setting various configuration information for transmitting / receiving wireless signals, a process for various signal processing (e.g., channel encoding / decoding, modulation / demodulation, resource mapping / demapping, etc.), and a resource allocation process.
[0059] NR supports multiple numerologies or subcarrier spacings (SCS) to support various 5G services. For example, when the SCS is 15 kHz, it supports a wide area in traditional cellular bands; when the SCS is 30 kHz / 60 kHz, it supports dense-urban areas, lower latency, and wider carrier bandwidth; and when the SCS is 60 kHz or higher, it supports a bandwidth greater than 24.25 GHz to overcome phase noise.
[0060] The NR frequency band can be defined by two types of frequency ranges (FR1, FR2). The numerical values of the frequency ranges may change. For example, the two types of frequency ranges (FR1, FR2) may be as shown in Table 1 below. For convenience of explanation, among the frequency ranges used in the NR system, FR1 may mean "sub 6GHz range" and FR2 may mean "above 6GHz range" and may be referred to as Millimeter Wave (mmW).
[0061] Frequency Range Definition Frequency Range Subcarrier Spacing FR1 450 MHz - 6000 MHz 15, 30, 60 kHz FR2 24 250 MHz - 52600 MHz 60, 120, 240 kHz
[0062] As described above, the numerical values of the frequency range of the NR system may change. For example, FR1 may include a band of 410 MHz to 7125 MHz as shown in Table 2 below. That is, FR1 may include a frequency band of 6 GHz (or 5850, 5900, 5925 MHz, etc.) or higher. For example, the frequency band of 6 GHz (or 5850, 5900, 5925 MHz, etc.) or higher included within FR1 may include an unlicensed band. The unlicensed band may be used for various purposes, for example, for communication for vehicles (e.g., autonomous driving).
[0063] Frequency Range Definition Frequency Range Subcarrier Spacing FR1 4 10 MHz - 7 125 MHz 15, 30, 60 kHz FR2 24 250 MHz - 5 2600 MHz 60, 120, 240 kHz
[0064] Here, the wireless communication technology implemented in the wireless device of the present disclosure may include LTE, NR, and 6G, as well as NarrowBand IoT (NB-IoT) for low-power communication. For example, NB-IoT technology may be an example of Low Power Wide Area Network (LPWAN) technology and may be implemented according to standards such as LTE Cat NB1 and / or LTE Cat NB2, but is not limited to the names mentioned above. Additionally or generally, the wireless communication technology implemented in the wireless device of the present disclosure may perform communication based on LTE-M technology. For example, LTE-M technology may be an example of LPWAN technology and may be referred to by various names such as eMTC (enhanced MTC). For example, LTE-M technology may be implemented in at least one of various standards such as 1) LTE CAT 0, 2) LTE Cat M1, 3) LTE Cat M2, 4) LTE non-BL (Non-Bandwidth Limited), 5) LTE-MTC, 6) LTE MTC, and / or 7) LTE M, and is not limited to the names mentioned above. Additionally or generally, wireless communication technology implemented in the wireless device of the present disclosure may include at least one of ZigBee, Bluetooth, and / or LPWAN for low-power communication, and is not limited to the names mentioned above. For example, ZigBee technology may create Personal Area Networks (PANs) related to small / low-power digital communication based on various standards such as IEEE 802.15.4, and may be referred to by various names.
[0065] FIG. 2 shows an example of a wireless device to which an implementation of the present disclosure is applied.
[0066] In FIG. 2, the first wireless device (100) and / or the second wireless device (200) may be implemented in various forms depending on the use example / service. For example, {the first wireless device (100) and the second wireless device (200)} may correspond to at least one of {wireless devices (100a–100f) and base station (200)}, {wireless devices (100a–100f) and wireless devices (100a–100f)} and / or {base station (200) and base station (200)} of FIG. 1. The first wireless device (100) and / or the second wireless device (200) may be composed of various components, devices / parts and / or modules.
[0067] The first wireless device (100) may include at least one transceiver such as a transceiver (106), at least one processing chip such as a processing chip (101), and / or one or more antennas (108).
[0068] The processing chip (101) may include at least one processor, such as a processor (102), and at least one memory, such as a memory (104). Additionally and / or generally, the memory (104) may be placed outside the processing chip (101).
[0069] The processor (102) can control the memory (104) and / or the transceiver (106) and may be configured to implement the description, function, procedure, proposal, method and / or operation flowchart disclosed in this disclosure. For example, the processor (102) may process information within the memory (104) to generate a first information / signal and transmit a wireless signal containing the first information / signal through the transceiver (106). The processor (102) may receive a wireless signal containing a second information / signal through the transceiver (106) and process the second information / signal to store the obtained information in the memory (104).
[0070] Memory (104) may be connected to the processor (102) so as to be operable. Memory (104) may store various types of information and / or instructions. Memory (104) may store firmware and / or software code (105) that implements code, instructions, and / or a set of instructions that perform the descriptions, functions, procedures, suggestions, methods, and / or operation flowcharts disclosed in this disclosure when executed by the processor (102). For example, the firmware and / or software code (105) may implement instructions that perform the descriptions, functions, procedures, suggestions, methods, and / or operation flowcharts disclosed in this disclosure when executed by the processor (102). For example, the firmware and / or software code (105) may control the processor (102) to perform one or more protocols. For example, the firmware and / or software code (105) may control the processor (102) to perform one or more wireless interface protocol layers.
[0071] Here, the processor (102) and memory (104) may be part of a communication modem / circuit / chip designed to implement RAT (e.g., LTE or NR). A transceiver (106) may be connected to the processor (102) and may transmit and / or receive a wireless signal through one or more antennas (108). Each transceiver (106) may include a transmitter and / or receiver. The transceiver (106) may be interchangeably used with an RF (Radio Frequency) unit. In this disclosure, the first wireless device (100) may represent a communication modem / circuit / chip.
[0072] The second wireless device (200) may include at least one transceiver such as a transceiver (206), at least one processing chip such as a processing chip (201), and / or one or more antennas (208).
[0073] The processing chip (201) may include at least one processor, such as a processor (202), and at least one memory, such as a memory (204). Additionally and / or alternatively, the memory (204) may be placed outside the processing chip (201).
[0074] The processor (202) can control the memory (204) and / or the transceiver (206) and may be configured to implement the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure. For example, the processor (202) may process information within the memory (204) to generate a third information / signal and transmit a wireless signal containing the third information / signal through the transceiver (206). The processor (202) may receive a wireless signal containing a fourth information / signal through the transceiver (206) and process the fourth information / signal to store the obtained information in the memory (204).
[0075] Memory (204) may be connected to the processor (202) so as to be operable. Memory (204) may store various types of information and / or instructions. Memory (204) may store firmware and / or software code (205) that implements code, instructions, and / or a set of instructions that perform the descriptions, functions, procedures, suggestions, methods, and / or operation flowcharts disclosed in this disclosure when executed by the processor (202). For example, the firmware and / or software code (205) may implement instructions that perform the descriptions, functions, procedures, suggestions, methods, and / or operation flowcharts disclosed in this disclosure when executed by the processor (202). For example, the firmware and / or software code (205) may control the processor (202) to perform one or more protocols. For example, the firmware and / or software code (205) may control the processor (202) to perform one or more wireless interface protocol layers.
[0076] Here, the processor (202) and memory (204) may be part of a communication modem / circuit / chip designed to implement a RAT (e.g., LTE or NR). A transceiver (206) may be connected to the processor (202) and may transmit and / or receive a wireless signal through one or more antennas (208). Each transceiver (206) may include a transmitter and / or receiver. The transceiver (206) may be interchangeably used with an RF unit. In this disclosure, the second wireless device (200) may represent a communication modem / circuit / chip.
[0077] Hereinafter, hardware elements of the wireless device (100, 200) will be described in more detail. Although not limited thereto, one or more protocol layers may be implemented by one or more processors (102, 202). For example, one or more processors (102, 202) may implement one or more layers (e.g., functional layers such as a PHY (physical) layer, a MAC (Media Access Control) layer, an RLC (Radio Link Control) layer, a PDCP (Packet Data Convergence Protocol) layer, an RRC (Radio Resource Control) layer, and an SDAP (Service Data Adaptation Protocol) layer). One or more processors (102, 202) may generate one or more PDUs (Protocol Data Units), one or more SDUs (Service Data Units), messages, control information, data, or information according to the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure. One or more processors (102, 202) may generate a signal (e.g., baseband signal) including a PDU, SDU, message, control information, data, or information according to the description, function, procedure, proposal, method, and / or operation flowchart disclosed in the present disclosure and provide it to one or more transceivers (106, 206). One or more processors (102, 202) may receive a signal (e.g., baseband signal) from one or more transceivers (106, 206) and may obtain a PDU, SDU, message, control information, data, or information according to the description, function, procedure, proposal, method, and / or operation flowchart disclosed in the present disclosure.
[0078] One or more processors (102, 202) may be referred to as a controller, a microcontroller, a microprocessor, and / or a microcomputer. One or more processors (102, 202) may be implemented by hardware, firmware, software, and / or a combination thereof. For example, one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), one or more Digital Signal Processing Devices (DSPDs), one or more Programmable Logic Devices (PLDs), and / or one or more Field Programmable Gate Arrays (FPGAs) may be included in one or more processors (102, 202). For example, one or more processors (102, 202) may be composed of a set of communication control processors, application processors (APs), electronic control units (ECUs), central processing units (CPUs), graphic processing units (GPUs), and memory control processors. One or more memories (104, 204) may be connected to one or more processors (102, 202) and may store various forms of data, signals, messages, information, programs, codes, instructions, and / or commands. One or more memories (104, 204) may be composed of Random Access Memory (RAM), Dynamic RAM (DRAM), Read-Only Memory (ROM), Erasable Programmable ROM (EPROM), flash memory, volatile memory, non-volatile memory, hard drive, register, cache memory, computer read storage media, and / or combinations thereof.One or more memories (104, 204) may be located inside and / or outside of one or more processors (102, 202). Additionally, one or more memories (104, 204) may be connected to one or more processors (102, 202) through various technologies such as wired or wireless connections.
[0079] One or more transceivers (106, 206) may transmit user data, control information, wireless signals / channels, etc., as described in the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure to one or more other devices. One or more transceivers (106, 206) may receive user data, control information, wireless signals / channels, etc., as described in the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure from one or more other devices. For example, one or more transceivers (106, 206) may be connected to one or more processors (102, 202) and may transmit and receive wireless signals. For example, one or more processors (102, 202) may control one or more transceivers (106, 206) to transmit user data, control information, wireless signals, etc., to one or more other devices. Additionally, one or more processors (102, 202) can control one or more transceivers (106, 206) to receive user data, control information, wireless signals, etc. from one or more other devices.
[0080] One or more transceivers (106, 206) may be connected to one or more antennas (108, 208). Additionally and / or generally, one or more transceivers (106, 206) may include one or more antennas (108, 208). One or more transceivers (106, 206) may be configured to transmit and receive user data, control information, wireless signals / channels, etc., mentioned in the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure through one or more antennas (108, 208). In this disclosure, one or more antennas (108, 208) may be a plurality of physical antennas or a plurality of logical antennas (e.g., antenna ports).
[0081] One or more transceivers (106, 206) can convert received user data, control information, wireless signals / channels, etc. from RF band signals to baseband signals in order to process received user data, control information, wireless signals / channels, etc. using one or more processors (102, 202). One or more transceivers (106, 206) can convert processed user data, control information, wireless signals / channels, etc. from baseband signals to RF band signals using one or more processors (102, 202). To this end, one or more transceivers (106, 206) may include (analog) oscillators and / or filters. For example, one or more transceivers (106, 206) can up-convert an OFDM baseband signal into an OFDM signal through an (analog) oscillator and / or filter under the control of one or more processors (102, 202) and transmit the up-converted OFDM signal at a carrier frequency. One or more transceivers (106, 206) can receive an OFDM signal at a carrier frequency and down-convert the OFDM signal into an OFDM baseband signal through an (analog) oscillator and / or filter under the control of one or more processors (102, 202).
[0082] Although not illustrated in FIG. 2, the wireless device (100, 200) may include additional components. The additional components (140) may be configured in various ways depending on the type of the wireless device (100, 200). For example, the additional components (140) may include at least one of a power unit / battery, an input / output (I / O) device (e.g., audio I / O port, video I / O port), a driving unit, and a computing unit. The additional components (140) may be connected to one or more processors (102, 202) through various technologies, such as wired or wireless connections.
[0083] In an embodiment of the present disclosure, the UE may operate as a transmitting device in the uplink and as a receiving device in the downlink. In an embodiment of the present disclosure, the base station may operate as a receiving device in the UL and as a transmitting device in the DL. For technical convenience, it is generally assumed that the first wireless device (100) operates as a UE and the second wireless device (200) operates as a base station. For example, a processor (102) connected to, mounted on, or released to the first wireless device (100) may be configured to perform UE operations according to an embodiment of the present disclosure or to control a transceiver (106) to perform UE operations according to an embodiment of the present disclosure. A processor (202) connected to, mounted on, or released to the second wireless device (200) may be configured to perform base station operations according to an embodiment of the present disclosure or to control a transceiver (206) to perform base station operations according to an embodiment of the present disclosure.
[0084] In the present disclosure, the base station may be referred to as Node B, eNode B, or gNB.
[0085] FIG. 3 shows an example of a UE to which an implementation of the present disclosure is applied.
[0086] Referring to FIG. 3, the UE (100) can correspond to the first wireless device (100) of FIG. 2.
[0087] The UE (100) includes a processor (102), memory (104), transceiver (106), one or more antennas (108), a power management module (141), a battery (142), a display (143), a keypad (144), a SIM (Subscriber Identification Module) card (145), a speaker (146), and a microphone (147).
[0088] The processor (102) may be configured to implement the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure. The processor (102) may be configured to control one or more other components of the UE (100) to implement the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this disclosure. Layers of a wireless interface protocol may be implemented in the processor (102). The processor (102) may include an ASIC, other chipsets, logic circuits, and / or data processing devices. The processor (102) may be an application processor. The processor (102) may include at least one of a DSP, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and a modem (modulator and demodulator). An example of the processor (102) is the SNAPDRAGON manufactured by Qualcomm®. TM Series processor, EXYNOS made by Samsung® TM Series processors, A Series processors made by Apple®, HELIO made by MediaTek® TM Series processors, ATOM made by Intel® TM It can be found in series processors or corresponding next-generation processors.
[0089] Memory (104) is coupled to the processor (102) so as to be operable and stores various information for operating the processor (102). Memory (104) may include ROM, RAM, flash memory, memory card, storage medium and / or other storage device. When the implementation is implemented in software, the technology described herein may be implemented using modules (e.g., procedures, functions, etc.) that perform the descriptions, functions, procedures, proposals, methods and / or operation flowcharts disclosed in this disclosure. Modules may be stored in memory (104) and executed by the processor (102). Memory (104) may be implemented within the processor (102) or outside the processor (102), in which case it may be communicatively coupled to the processor (102) through various methods known in the technology.
[0090] A transceiver (106) is coupled to operate with a processor (102) and transmits and / or receives a wireless signal. The transceiver (106) includes a transmitter and a receiver. The transceiver (106) may include a baseband circuit for processing a wireless frequency signal. The transceiver (106) controls one or more antennas (108) to transmit and / or receive a wireless signal.
[0091] The power management module (141) manages the power of the processor (102) and / or the transceiver (106). The battery (142) supplies power to the power management module (141).
[0092] The display (143) outputs the result processed by the processor (102). The keypad (144) receives input to be used by the processor (102). The keypad (144) can be displayed on the display (143).
[0093] A SIM card (145) is an integrated circuit for securely storing an International Mobile Subscriber Identity (IMSI) and associated keys, and is used to identify and authenticate a subscriber in a mobile device such as a mobile phone or computer. Additionally, contact information can be stored on many SIM cards.
[0094] The speaker (146) outputs sound-related results processed by the processor (102). The microphone (147) receives sound-related input to be used by the processor (102).
[0095] FIG. 4 shows an example of a frame structure in a 3GPP-based wireless communication system to which an implementation of the present disclosure is applied.
[0096] The frame structure illustrated in FIG. 4 is purely exemplary, and the number of subframes, the number of slots, and / or the number of symbols within the frame may vary. In a 3GPP-based wireless communication system, OFDM numerals (e.g., Sub-Carrier Spacing (SCS), Transmission Time Interval (TTI) durations) may be set differently among multiple aggregated cells for a single UE. For example, if a UE is set to different SCS for aggregated cells, the (absolute time) duration of time resources (e.g., subframes, slots, or TTI) containing the same number of symbols may differ between aggregated cells. Here, the symbols may include OFDM symbols (or CP-OFDM symbols) or SC-FDMA symbols (or DFT-s-OFDM (Discrete Fourier Transform-Spread-OFDM) symbols).
[0097] Referring to Fig. 4, downlink and uplink transmissions consist of frames. Each frame is T f= has a duration of 10ms. Each frame is divided into two half-frames, and the duration of each half-frame is 5ms. Each half-frame consists of 5 subframes, with a duration T per subframe. sf ε is 1ms. Each subframe is divided into slots, and the number of slots in a subframe depends on the subcarrier spacing. Each slot contains 14 or 12 OFDM symbols based on the Cyclic Prefix (CP). In a standard CP, each slot contains 14 OFDM symbols, and in an extended CP, each slot contains 12 OFDM symbols. Numericality is an exponentially expandable subcarrier spacing Δf = 2 u * Based on 15kHz.
[0098] Table 3 shows the subcarrier spacing Δf = 2 u * Number of OFDM symbols per slot N for a standard CP according to 15kHz slot symb , number of slots per frame N frame,u slot and the number of slots per subframe N subframe,u slot It represents.
[0099] uN slot symb N frame,u slot N subframe,u slot 01410111420221440431480841416016
[0100] Table 4 shows the subcarrier spacing Δf = 2 u * Number of OFDM symbols per slot for extended CP N according to 15kHz slot symb , number of slots per frame N frame,u slot and the number of slots per subframe N subframe,u slot It represents.
[0101] uN slotsymb N frame,u slot N subframe,u slot 212404
[0102] A slot contains multiple symbols (e.g., 14 or 12 symbols) in the time domain. For each numerology (e.g., subcarrier interval) and carrier, a Common Resource Block (CRB) N is represented by upper-layer signaling (e.g., RRC signaling). start,u grid N starting from size,u grid,x * N RB sc Subcarrier and N subframe,u symb The resource grid of OFDM symbols is defined. Here, N size,u grid,x is the number of Resource Blocks (RB) in the resource grid, and the subscript x is DL for downlinks and UL for uplinks. N RB sc is the number of subcarriers per RB. In 3GPP-based wireless communication systems, N RB sc is typically 12. There is one resource grid for a given antenna port p, subcarrier spacing setting u, and transmission direction (DL or UL). Carrier bandwidth N for subcarrier spacing setting u. size,u grid is given by upper-level parameters (e.g., RRC parameters). Each element of the resource grid for antenna port p and subcarrier spacing setting u is called a Resource Element (RE), and one complex symbol can be mapped to each RE. Each RE in the resource grid is uniquely identified by an index k in the frequency domain and an index l representing the symbol position relative to a reference point in the time domain. In 3GPP-based radio communication systems, an RB is defined as 12 consecutive subcarriers in the frequency domain.
[0103] In the 3GPP NR system, RBs are divided into CRBs and PRBs (Physical Resource Blocks). CRBs are numbered in the frequency domain starting from 0 and increasing for a subcarrier spacing setting u. The center of subcarrier 0 of CRB 0 for a subcarrier spacing setting u coincides with 'Point A', which serves as a common reference point for the resource block grid. In the 3GPP NR system, PRBs are defined within the Bandwidth Part (BWP) and are numbered from 0 to NsizeBWP,i-1, where i is the BWP number. The relationship between the PRB nPRB and CRB nCRB of BWP i is as follows: nPRB = nCRB + NsizeBWP,i, where NsizeBWP,i is the CRB for which the BWP starts relative to CRB 0. A BWP contains multiple consecutive RBs. A carrier can contain up to N (e.g., 5) BWPs. A UE can be configured with one or more BWPs on a given element carrier. Only one BWP can be active at a time among the BWPs configured on the UE. The active BWP defines the operating bandwidth of the UE within the operating bandwidth of the cell.
[0104] In the present disclosure, the term “cell” may mean a geographical area where one or more nodes provide a communication system, or it may mean a radio resource. A “cell” as a geographical area may be understood as coverage where a node can provide services using a carrier, and a “cell” as a radio resource (e.g., time-frequency resource) is associated with a bandwidth, which is a frequency range set by the carrier. A “cell” associated with a radio resource is defined as a combination of downlink resources and uplink resources, for example, a combination of a DL CC (Component Carrier) and an UL CC. A cell may consist only of downlink resources, or it may consist of downlink resources and uplink resources. Since DL coverage, which is the range where a node can transmit a valid signal, and UL coverage, which is the range where a node can receive a valid signal from a UE, depend on the carrier carrying the signal, the coverage of a node may be associated with the coverage of the “cell” of the radio resources used by the node. Therefore, the term "cell" may sometimes be used to indicate the service coverage of a node, at other times to indicate a wireless resource, or at other times to indicate the range over which a signal using the wireless resource can reach with effective strength.
[0105] In CA, two or more CCs are aggregated. Depending on its capabilities, the UE can receive or transmit simultaneously on one or more CCs. CA is supported for both consecutive and discontinuous CCs. When CA is configured, the UE has only one RRC connection with the network. During RRC connection establishment / re-establishment / handover, one serving cell provides NAS mobility information, and during RRC connection re-establishment / handover, one serving cell provides security input. This cell is called a PCell (Primary Cell). A PCell is a cell operating on the primary frequency where the UE performs the initial connection establishment procedure or initiates the connection re-establishment procedure. Depending on the UE's capabilities, a SCell (Secondary Cell) may be configured to form a set of serving cells together with the PCell. A SCell is a cell that provides additional radio resources on top of a Special Cell (SpCell). Therefore, the set of serving cells configured for a UE always consists of one PCell and one or more SCells. In the case of Dual Connectivity (DC) operation, the term SpCell refers to a PCell of a Master Cell Group (MCG) or a primary SCell (PSCell) of a Secondary Cell Group (SCG). SpCells support Physical Uplink Control Channel (PUCCH) transmission and contention-based random access, and are always active. An MCG is a group of serving cells associated with a master node, consisting of a SpCell (PCell) and optionally one or more SCells. An SCG is a group of serving cells associated with a secondary node, consisting of a PSCell and zero or more SCells, for a UE configured as a DC. For a UE in RRC_CONNECTED that is not configured as a CA / DC, only one serving cell consisting of a PCell exists.For a UE with RRC_CONNECTED configured as CA / DC, the term "serving cell" is used to refer to a set of cells consisting of SpCell(s) and all SCells. Two MAC objects are configured in the UE from the DC: one for the MCG and the other for the SCG.
[0106] This explains the Beam Management (BM) procedure.
[0107] A BM procedure is an L1 (layer 1) / L2 (layer 2) procedure for acquiring and maintaining a set of base station (e.g., gNB, TRP (Transmission Reception Point), etc.) and / or terminal (e.g., UE) beams that can be used for DL and / or UL transmission / reception, and may include the following procedures and terms.
[0108] - Beam measurement: An operation in which a base station or UE measures the characteristics of a received beamforming signal.
[0109] - Beam determination: The operation in which a base station or UE selects its transmit beam (Tx beam) / receive beam (Rx beam).
[0110] - Beam sweeping: An operation that covers a spatial area using transmitting and / or receiving beams for a set time interval in a predetermined manner.
[0111] - Beam report: An operation in which the UE reports information about the beam-formed signal based on beam measurements.
[0112] The BM procedure may include i) a DL BM procedure using an SS (Synchronization Signal) / PBCH (Physical Broadcast Channel) block (SSB) or CSI-RS (Channel State Information Reference Signal), and ii) a UL BM procedure using an SRS (Sounding Reference Signal).
[0113] Each BM procedure may include transmit beam sweeping to determine the transmit beam and receive beam sweeping to determine the receive beam.
[0114] The DL BM procedure may include i) transmission of the base station's beamformed DL RS (e.g., SSB or CSI-RS) and ii) beam reporting of the UE.
[0115] The beam report may include a preferred DL RS ID (Identifier) and / or a corresponding L1-RSRP (Reference Signal Received Power). The DL RS ID may include at least one of an SSBRI (SSB Resource Indicator) or a CRI (CSI-RS Resource Indicator).
[0116] FIG. 5 shows an example of DL RS transmission of a base station to which an implementation of the present disclosure is applied.
[0117] Referring to Fig. 5, an SSB beam and a CSI-RS beam can be used for BM. The measurement metric can be L1-RSRP per resource / block. The SSB is used for coarse BM, and the CSI-RS can be used for fine BM. The SSB can be used for both transmit beam sweeping and receive beam sweeping.
[0118] Receive beam sweeping using SSBs can be performed with the UE changing the receive beam for the same SSBRI over one or more SSB bursts. Here, one SSB burst includes one or more SSBs, and one set of SSB bursts includes one or more SSB bursts.
[0119] FIG. 6 shows an example of a DL BM procedure using an SSB to which an implementation of the present disclosure is applied.
[0120] Configuration for beam reporting using SSB can be performed during CSI / beam setup in the RRC connection state (e.g., RRC_CONNECTED).
[0121] In step S610, the UE receives from the base station a CSI-ResourceConfig IE containing a CSI-SSB-ResourceSetList containing SSB resources used for the BM.
[0122] Table 5 shows an example of CSI-ResourceConfig IE. For example, the BM configuration using SSB is not defined separately, and the SSB can be configured as a CSI-RS resource.
[0123] -- ASN1START-- TAG-CSI-RESOURCECONFIG-STARTCSI-ResourceConfig ::= SEQUENCE {csi-ResourceConfigId CSI-ResourceConfigId,csi-RS-ResourceSetList CHOICE {nzp-CSI-RS-SSB SEQUENCE {nzp-CSI-RS-ResourceSetList SEQUENCE (SIZE (1..maxNrofNZP-CSI-RS-ResourceSetsPerConfig)) OF NZP-CSI-RS-ResourceSetId OPTIONAL,csi-SSB-ResourceSetList SEQUENCE (SIZE (1..maxNrofCSI-SSB-ResourceSetsPerConfig)) OF CSI-SSB-ResourceSetId OPTIONAL},csi-IM-ResourceSetList SEQUENCE (SIZE (1..maxNrofCSI-IM-ResourceSetsPerConfig)) OF CSI-IM-ResourceSetId},bwp-Id BWP-Id,resourceType ENUMERATED { aperiodic, semiPersistent, periodic},...}-- TAG-CSI-RESOURCECONFIGTOADDMOD-STOP-- ASN1STOP
[0124] In Table 5, the csi-SSB-ResourceSetList IE (information element) can represent a list of SSB resources used for beam management and reporting in a single CSI-RS resource set. Here, the SSB resource set can be set to {SSBx1, SSBx2, SSBx3, SSBx4}. For example, the SSB index can be defined from 0 to 63.
[0125] In step S620, the UE receives an SSB resource based on the CSI-SSB-ResourceSetList from the base station.
[0126] In step S630, if a CSI-ReportConfig related to reporting for SSBRI and L1-RSRP is configured, the UE reports the best SSBRI and / or the corresponding L1-RSRP to the base station (S430). For example, if the reportQuantity of the CSI-ReportConfig IE is set to 'ssb-Index-RSRP', the UE may report the best SSBRI and / or the corresponding L1-RSRP to the base station.
[0127] Additionally, if the UE establishes a CSI-RS resource on the same OFDM symbol as the SSB and 'QCL-TypeD' is applicable, the UE may assume that the CSI-RS and SSB are in a Quasi-Co-Located (QCL) relationship in terms of 'QCL-TypeD'. QCL-TypeD may imply that antenna ports are QCL-colocated in terms of spatial reception parameters. When the UE receives signals through multiple DL antenna ports in a QCL-TypeD relationship, the same reception beam may be applied. Furthermore, the UE does not expect the CSI-RS to be established on an RE that overlaps with the RE of the SSB.
[0128] CSI-RS can be used for various purposes. If the repetition parameter is set for a specific set of CSI-RS resources and trs-Info is not set, CSI-RS can be used for beam management. If the repetition parameter is not set for a specific set of CSI-RS resources and trs-Info is set, CSI-RS can be used for Tracking Reference Signal (TRS). If the repetition parameter is not set for a specific set of CSI-RS resources and trs-Info is not set, CSI-RS can be used for CSI acquisition. The repetition parameter can only be set for CSI-RS resources associated with a CSI-ReportConfig set to L1-RSRP or 'No Report (or None)' reporting.
[0129] If a UE receives a CSI-ReportConfig with reportQuantity set to 'cri-RSRP', 'cri-SINR', or 'none', and if the CSI-ResourceConfig for channel measurement (by the upper-level parameter resourcesForChannelMeasurement) contains an NZP-CSI-RS-ResourceSet with 'repetition' set but not containing 'trs-Info', the UE can only set all CSI-RS resources within the NZP-CSI-RS-ResourceSet to 'nrofPorts' with the same number of ports (1-port or 2-port).
[0130] Setting the repetition parameter to 'ON' may be associated with receive beam sweeping. When the repetition parameter is set to 'ON', the UE can assume that at least one CSI-RS resource within the NZP-CSI-RS-ResourceSet is transmitted through the same DL spatial domain transmission filter. That is, at least one CSI-RS resource within the NZP-CSI-RS-ResourceSet is transmitted through the same transmit beam. Here, at least one CSI-RS resource within the NZP-CSI-RS-ResourceSet may be transmitted with different OFDM symbols. Additionally, the UE does not expect to receive different periods in periodicityAndOffset from all CSI-RS resources within the NZP-CSI-RS-ResourceSet.
[0131] On the other hand, setting the repetition parameter to 'OFF' may be associated with the base station's transmit beam sweeping. When the repetition parameter is set to 'OFF', the UE does not assume that at least one CSI-RS resource within NZP-CSI-RS-ResourceSet is transmitted via the same DL spatial domain transmission filter. That is, at least one CSI-RS resource within NZP-CSI-RS-ResourceSet is transmitted via a different transmit beam.
[0132] FIG. 7 shows an example of a DL BM procedure using CSI-RS to which an implementation of the present disclosure is applied.
[0133] FIG. 7-(a) illustrates the receiving beam determination (or beam refinement) procedure of the UE. FIG. 7-(b) illustrates the transmitting beam sweeping procedure of the base station. Additionally, FIG. 7-(a) is the case where the repetition parameter is set to 'ON', and FIG. 7-(b) is the case where the repetition parameter is set to 'OFF'.
[0134] FIG. 8 shows an example of a receiving beam determination procedure of a UE to which an implementation of the present disclosure is applied.
[0135] In step S810, the UE receives an NZP CSI-RS resource set IE containing an upper layer parameter, a repeating parameter, from the base station via RRC signaling. The repeating parameter is set to 'ON'.
[0136] In step S820, the UE repeatedly receives CSI-RS resources within an NZP CSI-RS resource set with the repetition parameter set to 'ON' in different OFDM symbols through the same transmit beam (or DL spatial domain transmit filter) of the base station.
[0137] In step S830, the UE determines its receiving beam.
[0138] In step S840, the UE skips the CSI report. In this case, the reportQuantity of the CSI report settings can be set to 'No report (or None)'.
[0139] In other words, the UE can skip CSI reporting if the repetition parameter is set to 'ON'.
[0140] FIG. 9 shows an example of a transmission beam determination procedure of a base station to which an implementation of the present disclosure is applied.
[0141] In step S910, the UE receives an NZP CSI-RS resource set IE containing an upper layer parameter, a repetition parameter, from the base station via RRC signaling. The repetition parameter is set to 'OFF' and is associated with the base station's transmit beam sweeping procedure.
[0142] In step S920, the UE receives CSI-RS resources within an NZP CSI-RS resource set with the repetition parameter set to 'OFF' through a different transmit beam (or DL spatial domain transmit filter) of the base station.
[0143] In step S930, the UE selects / determines the best beam.
[0144] In step S940, the UE reports the ID and / or related quality information (e.g., L1-RSRP) for the selected beam to the base station. In this case, the reportQuantity of the CSI reporting settings can be set to 'CRI + L1-RSRP'.
[0145] That is, when CSI-RS is transmitted for BM, the UE can report CRI and L1-RSRP for it to the base station.
[0146] Depending on the UE implementation, beam reciprocity or beam correspondence may or may not exist between the transmit beam and the receive beam in the UL BM procedure. If beam reciprocity exists between the transmit beam and the receive beam at both the base station and the UE, the UL beam pair can be matched through the DL beam pair. However, if beam reciprocity does not exist between the transmit beam and the receive beam at either the base station or the UE, the UL beam pair determination may be required separately from the DL beam pair determination. Additionally, even if both the base station and the UE maintain beam reciprocity, the base station may use the UL BM procedure to determine the DL transmit beam without the UE requesting a report of the preferred beam.
[0147] The UL BM procedure can be performed through beamformed UL SRS transmission. Whether UL BM is applied to an SRS resource set can be set by the usage parameter, which is a higher-level parameter. If the usage parameter is set to 'BeamManagement', only one SRS resource can be transmitted for each of multiple SRS resource sets in a given time instant. However, SRS resources within different SRS resource sets that have the same time domain operation within the same BWP can be transmitted simultaneously.
[0148] A UE may be configured with one or more SRS resources set by the upper-level parameter SRS-ResourceSet or SRS-PosResourceSet-r16. For each set of SRS resources set by SRS-ResourceSet, the UE may be configured with K≥1 SRS resources (by the upper-level parameter SRS-resource). Here, K is a natural number, and the maximum value of K may be indicated by SRS_capability. When an SRS is configured by SRS-PosResourceSet-r16, K SRS resources may be configured for the UE, and the maximum value of K may be 16.
[0149] Similar to the DL BM procedure, the UL BM procedure may also include transmit beam sweeping of the UE and receive beam sweeping of the base station.
[0150] FIG. 10 shows an example of a UL BM procedure using an SRS to which an implementation of the present disclosure is applied.
[0151] Figure 10-(a) shows the receiving beam determination procedure of a base station. Figure 10-(b) shows the transmitting beam sweeping procedure of a UE.
[0152] FIG. 11 shows an example of a transmit beam determination procedure of a UE to which an implementation of the present disclosure is applied.
[0153] In step S1110, the UE receives RRC signaling (e.g., SRS-Config IE) from the base station, which includes a usage parameter that is a higher-level parameter set to 'beam management'.
[0154] Table 6 shows an example of an SRS-Config IE. An SRS-Config IE is used to configure SRS transmission settings and / or SRS measurements for Cross-Link Interference (CLI). An SRS-Config IE may include a list of SRS-Resources and a list of SRS-ResourceSets. Each set of SRS resources may represent a set of SRS-resources.
[0155] The network can trigger the transmission of an SRS resource set using a configured aperiodicSRS-ResourceTrigger (L1 DCI).
[0156] -- ASN1START-- TAG-MAC-CELL-GROUP-CONFIG-STARTSRS-Config ::= SEQUENCE {srs-ResourceSetToReleaseList SEQUENCE (SIZE(1..maxNrofSRS-ResourceSets)) OF SRS-ResourceSetId OPTIONAL, -- Need Nsrs-ResourceSetToAddModList SEQUENCE (SIZE(1..maxNrofSRS-ResourceSets)) OF SRS-ResourceSet OPTIONAL, -- Need Nsrs-ResourceToReleaseList SEQUENCE (SIZE(1..maxNrofSRS-Resources)) OF SRS-ResourceId OPTIONAL, -- Need Nsrs-ResourceToAddModList SEQUENCE (SIZE(1..maxNrofSRS-Resources)) OF SRS-Resource OPTIONAL, -- Need Ntpc-Accumulation ENUMERATED {disabled} OPTIONAL, -- Need S...}SRS-ResourceSet ::= SEQUENCE {srs-ResourceSetId SRS-ResourceSetId,srs-ResourceIdList SEQUENCE (SIZE(1..maxNrofSRS-ResourcesPerSet)) OF SRS-ResourceId OPTIONAL, -- Cond SetupresourceType CHOICE {aperiodic SEQUENCE {aperiodicSRS-ResourceTrigger INTEGER (1..maxNrofSRS-TriggerStates-1),csi-RS NZP-CSI-RS-ResourceId OPTIONAL, -- Cond NonCodebookslotOffset INTEGER (1..32) OPTIONAL, -- Need S...,[[aperiodicSRS-ResourceTriggerList SEQUENCE (SIZE(1..maxNrofSRS-TriggerStates-2))OF INTEGER (1..maxNrofSRS-TriggerStates-1) OPTIONAL -- Need M]]},semi-persistent SEQUENCE {associatedCSI-RS NZP-CSI-RS-ResourceId OPTIONAL, -- Cond NonCodebook...},periodic SEQUENCE {associatedCSI-RS NZP-CSI-RS-ResourceId OPTIONAL, -- Cond NonCodebook...}},usage ENUMERATED {beamManagement, codebook, nonCodebook, antennaSwitching},alpha Alpha OPTIONAL, -- Need Sp0 INTEGER (-202..24) OPTIONAL, -- Cond SetuppathlossReferenceRS PathlossReferenceRS-Config OPTIONAL, -- Need Msrs-PowerControlAdjustmentStates ENUMERATED { sameAsFci2, separateClosedLoop} OPTIONAL, -- Need S...,[[pathlossReferenceRSList-r16 SetupRelease { PathlossReferenceRSList-r16} OPTIONAL -- Need M]]}PathlossReferenceRS-Config ::= CHOICE {ssb-Index SSB-Index,csi-RS-Index NZP-CSI-RS-ResourceId}SRS-PosResourceSet-r16 ::= SEQUENCE {srs-PosResourceSetId-r16 SRS-PosResourceSetId-r16,srs-PosResourceIdList-r16 SEQUENCE (SIZE(1..maxNrofSRS-ResourcesPerSet)) OF SRS-PosResourceId-r16OPTIONAL, -- Cond SetupresourceType-r16 CHOICE {aperiodic-r16 SEQUENCE {aperiodicSRS-ResourceTriggerList-r16 SEQUENCE (SIZE(1..maxNrofSRS-TriggerStates-1))OF INTEGER (1..maxNrofSRS-TriggerStates-1) OPTIONAL, -- Need M...},semi-persistent-r16 SEQUENCE {...},periodic-r16 SEQUENCE {...}},alpha-r16 Alpha OPTIONAL, -- Need Sp0-r16 INTEGER (-202..24) OPTIONAL, -- Cond SetuppathlossReferenceRS-Pos-r16 CHOICE {ssb-IndexServing-r16 SSB-Index,ssb-Ncell-r16 SSB-InfoNcell-r16,dl-PRS-r16 DL-PRS-Info-r16} OPTIONAL, -- Need M...}SRS-SpatialRelationInfo ::= SEQUENCE {servingCellId ServCellIndex OPTIONAL, -- Need SreferenceSignal CHOICE {ssb-Index SSB-Index,csi-RS-Index NZP-CSI-RS-ResourceId,srs SEQUENCE {resourceId SRS-ResourceId,uplinkBWP BWP-Id}}}SRS-SpatialRelationInfoPos-r16 ::= CHOICE {servingRS-r16 SEQUENCE {servingCellId ServCellIndex OPTIONAL, -- Need SreferenceSignal-r16 CHOICE {ssb-IndexServing-r16 SSB-Index,csi-RS-IndexServing-r16 NZP-CSI-RS-ResourceId,srs-SpatialRelation-r16 SEQUENCE {resourceSelection-r16 CHOICE {srs-ResourceId-r16 SRS-ResourceId,srs-PosResourceId-r16 SRS-PosResourceId-r16},uplinkBWP-r16 BWP-Id}}},ssb-Ncell-r16 SSB-InfoNcell-r16,dl-PRS-r16 DL-PRS-Info-r16}SRS-ResourceId ::= INTEGER (0..maxNrofSRS-Resources-1).
[0157] In Table 6, the usage parameter ('usage') represents a higher-level parameter indicating whether the SRS resource set is used for BM, or for codebook-based or non-codebook-based transmission. The usage parameter may correspond to the L1 parameter 'SRS-SetUse'. 'spatialRelationInfo' or 'spatialRelationInfoPos-r16' is a parameter indicating the setting of the spatial relation between the reference RS and the target SRS. Here, the reference RS can be the SSB, CSI-RS, or SRS corresponding to the L1 parameter 'SRS-SpatialRelationInfo'. If the SRS is set by SRS-PosResourceSet-r16, the reference RS can also be the DL PRS (Positioning Reference Signal). The usage parameter can be set per SRS resource set.
[0158] In step S1120, the UE determines the transmit beam for the SRS resource to be transmitted based on the spatialRelationInfo included in the SRS-Config IE (S1020). The spatialRelationInfo can be configured per SRS resource and may indicate whether to apply the same beam used in the SSB, CSI-RS, or SRS for each SRS resource. Additionally, the spatialRelationInfo may or may not be configured for each SRS resource.
[0159] In step S1130, if spatialRelationInfo is set in the SRS resource, the UE transmits by applying the same beam used in the SSB, CSI-RS, or SRS. However, if spatialRelationInfo is not set in the SRS resource, the UE arbitrarily determines a transmit beam and transmits the SRS through the determined transmit beam.
[0160] More specifically, for periodic SRSs where resourceType in SRS-Resource or SRS-PosResource-r16 is set to 'periodic':
[0161] i) If spatialRelationInfo or spatialRelationInfoPos-r16 is set to 'SSB / PBCH', the UE may transmit the SRS resource by applying a spatial domain transmission filter identical to (or generated from) the spatial domain receive filter used for receiving the SSB / PBCH; or
[0162] ii) If spatialRelationInfo or spatialRelationInfoPos-r16 is set to 'CSI-RS', the UE may transmit SRS resources by applying the same spatial domain transmission filter used for receiving CSI-RS; or
[0163] iii) If spatialRelationInfo or spatialRelationInfoPos-r16 is set to 'SRS', the UE may transmit the SRS resource by applying the same spatial domain transmission filter used for the transmission of periodic SRS; or
[0164] iv) When spatialRelationInfoPos-r16 is set to 'PRS', the UE can transmit the corresponding SRS resources by applying the same spatial domain transmission filter used for receiving the DL PRS.
[0165] Even if the resourceType within SRS-Resource or SRS-PosResource-r16 is set to 'SP-SRS' or 'AP-SRS', beam determination and transmission operations can be applied similarly to the above.
[0166] In step S1140, the UE may additionally receive feedback regarding the SRS from the base station. In this case, any one of the following three may apply.
[0167] i) When spatialRelationInfo is set for all SRS resources within an SRS resource set, the UE can transmit SRS to the beam indicated by the base station. For example, if spatialRelationInfo indicates the same SSB, CRI, or SRI, the UE can repeatedly transmit SRS to the same beam. This case corresponds to FIG. 10-(a) for the purpose of sweeping the receiving beam by the base station.
[0168] ii) SpatialRelationInfo may not be set for all SRS resources within the SRS resource set. In this case, the UE can freely switch SRS beams and transmit. This case corresponds to FIG. 10-(b) for the purpose of sweeping the transmission beam by the UE.
[0169] iii) SpatialRelationInfo may be set for only some SRS resources within a set of SRS resources. In this case, the UE may transmit SRS using the designated beam for SRS resources for which spatialRelationInfo is set, and transmit by applying an arbitrary transmission beam for SRS resources for which spatialRelationInfo is not set.
[0170] Advancements in AI (Artificial Intelligence) and ML (Machine Learning) technologies are leading to the intelligentization and sophistication of nodes and UEs that constitute wireless communication networks. In particular, due to the intelligence of networks and base stations, it is expected that various network and base station decision parameters (e.g., transmit / receive power of each base station, transmit power of each UE, precoders / beams of base stations and UEs, time / frequency resource allocation for each UE, duplex mode of each base station, etc.) will be rapidly optimized, derived, and applied in response to diverse environmental parameters (e.g., distribution / location of base stations, distribution / location / material of buildings / furniture, location / movement direction / speed of UEs, weather information, etc.).
[0171] AI / ML support for NG-RAN as an NG (Next Generation)-RAN (Radio Access Network) function is used to utilize AI / ML technology in NG-RAN.
[0172] The purpose of AI / ML for NG-RAN is to improve network performance and user experience by analyzing data collected and autonomously processed by NG-RAN, thereby enabling additional insights into, for example, network energy saving, load balancing, mobility optimization, network slicing, and Coverage and Capacity Optimization (CCO).
[0173] AI / ML support for NG-RAN may require input from adjacent NG-RAN nodes (e.g., prediction information, feedback information, measurements) and / or UEs (e.g., measurement results).
[0174] Signaling procedures used for information exchange for AI / ML support for NG-RAN are agnostic to use cases and data types, which may mean that the intended use (e.g., input, output, feedback) of the data exchanged through such procedures is not indicated.
[0175] The following scenarios can be supported for AI / ML deployment for NG-RAN:
[0176] - AI / ML model training is located in OAM (Operation, Administration, and Maintenance), and AI / ML model inference is located in the NG-RAN node;
[0177] - Both AI / ML model training and AI / ML model inference are located on the NG-RAN node.
[0178] AI / ML models must be trained, validated, and tested before deployment for AI / ML model inference.
[0179] The following definitions may be used in relation to AI / ML.
[0180] - ML model: A manageable representation of an ML model algorithm
[0181] - ML model training: A process performed by an ML training function to take training data, execute it through an ML model algorithm, derive a relevant loss, iteratively adjust the parameterization of the ML model based on the calculated loss, and generate a trained ML model.
[0182] - ML model initial training: The process of training an initial version of an ML model.
[0183] - ML model retraining: The process of training a previously trained version of an ML model and generating a new version.
[0184] - ML model pre-specialized training: The process of training an ML model with a dataset not specific to any inference type.
[0185] - ML model fine-tuning: The process of training a pre-specialized ML model to narrow the inference range of the ML model to a new single type of inference and create a new ML model.
[0186] - AI / ML Inference: The process of running an input dataset through a trained ML model to generate an output dataset, such as predictions.
[0187] - AI / ML inference function: A logical function that performs inference using a trained ML model.
[0188] - AI / ML Inference Emulation: Running the inference process in an emulation environment to evaluate the performance of an ML model before deploying it to a target environment.
[0189] - ML model deployment: The process of making a trained ML model available for use in a target environment.
[0190] - ML model loading: The process of making a trained ML model available for use in inference functions.
[0191] - AI / ML activation: The process of activating the inference function of the AI / ML inference function.
[0192] - AI / ML deactivation: The process of disabling the inference function of the AI / ML inference feature.
[0193] The following information can be configured to be reported by NG-RAN nodes:
[0194] - Predicted resource status information including predicted wireless resource status by slice;
[0195] - UE performance feedback including UE performance by slice;
[0196] - Measured UE trajectory;
[0197] - Energy Cost (EC);
[0198] - Predicted slice available capacity
[0199] The collection and reporting of the above information are established through the data collection reporting initiation procedure, and actual reporting can be performed through the data collection reporting procedure.
[0200] The collection of measured UE trajectories and UE performance feedback can be triggered upon a successful handover.
[0201] For example, cell-based UE trajectory prediction, which can be used in mobility optimization use cases, can be transmitted to a target NG-RAN node via a handover preparation procedure to provide information for subsequent mobility decisions, for example. Cell-based UE trajectory prediction may be limited to a first-hop target NG-RAN node.
[0202] An NG-RAN node can derive the corresponding future coverage status for predicted CCO issues and affected cells and beams. An NG-RAN node can notify adjacent NG-RAN nodes of the future coverage status change, along with correction cause and time information, through an NG-RAN node configuration update procedure. An NG-RAN node can also notify adjacent NG-RAN nodes that a previously notified coverage status change has been cancelled, along with the corresponding correction cause.
[0203] For a list of predicted affected cells and beams at any given point in time, there can be only one predicted CCO issue generated by gNB.
[0204] In addition, the application of AI / ML technology to the wireless interface (air interface) between the UE and the network is being discussed. Major use cases for applying AI / ML technology to the wireless interface may include BM, CSI acquisition / prediction, and positioning.
[0205] AI / ML-based beam management utilizes intracellular downlink beam prediction in serving cells to reduce measurement / RS overhead and improve beam selection accuracy. Two types of beam prediction can be supported:
[0206] - Spatial domain downlink transmission beam prediction for a beam set based on measurement results of another beam set, where the beam set for prediction may be an SSB or CSI-RS beam and the other beam set for measurement may be an SSB or CSI-RS beam; and
[0207] - Time-domain downlink transmission beam prediction for a beam set based on past measurement results of different beam sets, where the two beam sets may differ.
[0208] For AI / ML-based beam management, both network-side and UE-side models can be supported.
[0209] For UE-side models, the gNB can provide inference settings and / or inference-related parameters based on the features supported by the UE. The UE can report to the gNB applicable features, inapplicable features along with preferences for unsetting, and subsequent changes to the applicability status of the features.
[0210] For the UE-side model, the gNB can initiate performance monitoring and make management decisions based on the performance monitoring results. The UE can be configured to transmit either a measurement report or a calculated performance metric.
[0211] UE-side data collection can be applied to AI / ML beam management and CSI prediction functions. For UE-side data collection for UE-side model training, the gNB can configure whether to allow the UE to initiate requests for data collection settings (e.g., the UE's preference for starting or stopping data collection, preferred settings from the list of candidate settings provided by the network). The gNB can also provide or disable data collection settings to the UE at any time, regardless of a UE request.
[0212] FIG. 12 shows an example of a functional framework of AI / ML to which an implementation of the present disclosure is applied.
[0213] Referring to FIG. 12, the functional framework of AI / ML for an NR wireless interface may include a data collection function, a model training function, a model storage function, a management function, and an inference function. Among these, the management function can determine the behavior (e.g., selection / (de)activation / switching / fallback) of the AI / ML model and / or AI / function and monitor performance. Monitoring methods may include UE-side monitoring, network-side monitoring, and hybrid monitoring, depending on the entity calculating the monitoring metrics. The management function can play a role in making decisions to ensure appropriate inference behavior based on data received from the data collection function and the inference function.
[0214] Management instructions are information required as input for management functions to manage inference functions. This information may include the selection or (de)enablement of AI / ML models or functions, fallback to non-AI / ML behavior (i.e., not relying on the inference process), etc.
[0215] - A model transfer / delivery request is used to request a model from the model storage function.
[0216] Performance feedback / retraining requests are information required as input to the model training function, for example, for the purpose of model (re)training or updating.
[0217] When the performance of AI / ML models and / or AI / ML functions degrades, there are no appropriate criteria in the current management functions to determine subsequent replacement behavior. If there are no appropriate criteria to determine replacement behavior upon performance degradation, model management overhead may increase due to unnecessary Life Cycle Management (LCM) operations.
[0218] Furthermore, when the performance of AI / ML models and / or AI / ML functions degrades, the current management function does not consider dynamic situational changes. For example, when the performance of AI / ML models and / or AI / ML functions degrades, the current management function does not determine subsequent alternative actions by considering circumstances such as the UE's location or channel environment. Depending on dynamic situational changes, the optimal alternative action applicable on the UE or network side may vary.
[0219] To address the above problems, the present disclosure provides a method for improving the generalization performance of an AI / ML model by using an evaluation model that predicts the performance of alternative actions to subsequently determine an optimized alternative action.
[0220] More specifically, the present disclosure provides a method and procedure for determining an optimal alternative action by using an evaluation model that predicts the performance of an alternative action selected based on configuration and / or reporting by the network or the UE when the performance of a UE-side AI / ML model and / or AI / ML function degrades (e.g., when the performance of the primary model degrades). The network may determine the optimal alternative action. Monitoring metrics for determining the optimal alternative action may be calculated by the network or the UE. That is, the present disclosure may be applicable to UE-side monitoring, network-side monitoring, and hybrid monitoring.
[0221] In the present disclosure, the alternative behavior may be at least one of a candidate model and / or a pre-defined legacy mode in model switching. The pre-defined legacy mode may be a fallback mode. There may be one or more candidate models and pre-defined legacy modes. For example, there may be M candidate models and L legacy modes. Model switching may mean deactivating the currently active AI / ML model for a specific AI / ML support feature and enabling another AI / ML model. A fallback mode may mean changing the behavior from an AI / ML model and / or AI / ML function to a non-AI / ML-based legacy technique.
[0222] In the present disclosure, the evaluation model of an alternative action (or, performance prediction model) refers to an AI / ML model that predicts the performance of an alternative action and may be a proxy model and / or surrogate model possessed by the network.
[0223] The input to the evaluation model may include at least one of the data of the input to the main model, the output of the main model, or additional information (processing time, compression ratio, etc.) related to the inference of the main model. For example, the input to the evaluation model may include measurement statistics based on measurement RS and / or ground truth RS.
[0224] The output of the evaluation model is the performance of the alternative action, and the performance of the alternative action can be determined by at least one of the following performance indicators. The network may have multiple evaluation models for each alternative action that predict each performance indicator.
[0225] - Accuracy: For example, Mean Squared Error (MSE) with respect to the correct label
[0226] - Throughput: For example, link-level or system-level throughput)
[0227] - Overhead: For example, computational complexity of a specific alternative operation / memory and power usage / network resource utilization / signaling overhead
[0228] - Latency: For example, computation time of a specific legacy mode or inference time of a candidate model (i.e., the timeliness of monitoring results from the occurrence of a model error to the implementation of corrective action, considering the purpose of model monitoring), etc.
[0229] In the methods and / or procedures described in this disclosure, multiple steps may be performed simultaneously and / or in parallel. In the methods and / or procedures described in this disclosure, multiple steps may be performed in a different order than that described in the drawings. In the methods and / or procedures described in this disclosure, some steps may be omitted without loss of generality. The methods and / or procedures described in this disclosure may be performed or used in combination or complementarily.
[0230] The names of information, indications, and / or parameters described in this disclosure are merely illustrative. The names of information, indications, and / or parameters described in this disclosure may be replaced with other names or interpreted as such for the procedures, purposes, and methods proposed in this disclosure.
[0231] The following drawings are made to illustrate a specific example of the present disclosure. The names of specific devices or specific signals / messages / fields described in the drawings are presented as examples, and therefore the technical features of the present disclosure are not limited to the specific names used in the following drawings.
[0232] FIG. 13 illustrates an example of a method performed by a UE applicable to an implementation of the present disclosure.
[0233] In step S1310, the method includes a step of determining performance degradation of the operation.
[0234] In some implementations, the above operation may be a UE-side AI / ML-based operation.
[0235] In step S1320, the method includes the step of reporting to a network information related to one or more alternative actions applicable to the action. Performance prediction values of the one or more alternative actions are derived based on an evaluation model.
[0236] In some implementations, the method may further include the step of receiving configuration information related to one or more alternative operations from the network. The configuration information related to one or more alternative operations may include at least one of i) the type and / or category of an evaluation model currently available to the network, or ii) configuration information required for the operation of the evaluation model.
[0237] In some implementations, the input of the evaluation model is the same as the input of the model associated with the operation, and the output of the evaluation model may consist of performance prediction values of one or more alternative operations.
[0238] In some implementations, the above one or more alternative actions may be defined based on a set of alternative IDs (identifiers).
[0239] In some implementations, the method may further include a step of reporting information for selecting one or more alternative actions or types of evaluation models.
[0240] In step S1330, the method includes the step of receiving from the network configuration information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions.
[0241] In some implementations, the performance predictions of one or more alternative actions may be derived from the network based on an evaluation model.
[0242] In some implementations, performance predictions of one or more alternative actions may be derived from the UE based on an evaluation model. In this case, the method may further include the step of receiving configuration information related to the evaluation model from the network. Additionally, the method may further include the step of reporting the derived performance predictions to the network.
[0243] In some implementations, the optimal alternative action may include at least one of action switching, fallback, or reactivation of the action.
[0244] In some implementations, the configuration information related to the optimal alternative operation may include at least one of a condition for switching the operation, a condition for the fallback, or a condition for reactivating the operation.
[0245] In step S1340, the method includes the step of executing the optimal alternative action based on the setting information.
[0246] Additionally, the method described in Fig. 13 from the perspective of the UE may be performed by the first wireless device (100) shown in Fig. 2 and / or the UE (100) shown in Fig. 3.
[0247] The UE includes at least one transceiver, at least one processor, and at least one memory that stores instructions for the UE to perform the method described in FIG. 13 based on being connected to the at least one processor to be operable with the at least one processor and being executed by the at least one processor.
[0248] More specifically, the UE includes a step of determining performance degradation of the operation.
[0249] In some implementations, the above operation may be a UE-side AI / ML-based operation.
[0250] The UE reports information related to one or more alternative actions applicable to the above action to the network. Performance predictions of the one or more alternative actions are derived based on an evaluation model.
[0251] In some implementations, the UE may receive configuration information related to one or more alternative actions from the network. The configuration information related to one or more alternative actions may include at least one of i) the type and / or category of an evaluation model currently available to the network, or ii) configuration information required for the operation of the evaluation model.
[0252] In some implementations, the input of the evaluation model is the same as the input of the model associated with the operation, and the output of the evaluation model may consist of performance prediction values of one or more alternative operations.
[0253] In some implementations, the above one or more alternative actions may be defined based on a set of alternative IDs (identifiers).
[0254] In some implementations, the UE may report information for selecting one or more alternative behaviors or types of evaluation models.
[0255] In step S1330, the UE receives configuration information from the network related to the optimal alternative action selected based on the performance prediction values of one or more alternative actions.
[0256] In some implementations, the performance predictions of one or more alternative actions may be derived from the network based on an evaluation model.
[0257] In some implementations, performance predictions of one or more alternative actions may be derived from the UE based on an evaluation model. In this case, the UE may receive configuration information related to the evaluation model from the network. Additionally, the UE may report the derived performance predictions to the network.
[0258] In some implementations, the optimal alternative action may include at least one of action switching, fallback, or reactivation of the action.
[0259] In some implementations, the configuration information related to the optimal alternative operation may include at least one of a condition for switching the operation, a condition for the fallback, or a condition for reactivating the operation.
[0260] The UE executes the optimal alternative action based on the above configuration information.
[0261] Additionally, the method described in Fig. 13 from the perspective of the UE may be performed by the control of the processor (102) included in the first wireless device (100) shown in Fig. 2 and / or the control of the processor (102) included in the UE (100) shown in Fig. 3.
[0262] The device includes at least one processor integrated with the UE, and at least one memory storing processor-executable instructions configured to enable the at least one processor to perform the method described in FIG. 13.
[0263] Additionally, the method described in Fig. 13 from the perspective of the UE can be performed by software code (105) stored in the memory (104) included in the first wireless device (100) shown in Fig. 2.
[0264] The technical features of the present disclosure may be implemented directly in hardware, in software executed by a processor, or in a combination of both. For example, a method performed by a wireless device in wireless communication may be implemented in hardware, software, firmware, or a combination thereof. For example, the software may be in RAM, flash memory, ROM, EPROM, EEPROM, registers, a hard disk, a removable disk, a CD-ROM, or other storage media.
[0265] Some examples of storage media may be coupled to the processor so that the processor can read information from the storage media. Alternatively, the storage media may be integrated into the processor. The processor and storage media may exist in an ASIC. In other examples, the processor and storage media may exist as separate components.
[0266] Computer-readable media may include non-transitory computer-readable storage media of a type.
[0267] For example, non-transient computer-readable media may include RAM such as SDRAM (Synchronous Dynamic RAM), ROM, non-volatile NVRAM (Non-Volatile RAM), EEPROM, flash memory, magnetic or optical data storage media, or other media that can be used to store instructions or data structures. Non-transient computer-readable media may include combinations of the above.
[0268] Additionally, the method described in the present disclosure may be realized by a computer-readable communication medium that carries or communicates code in the form of at least partially instructions or data structures and can be accessed, read, and / or executed by a computer.
[0269] According to some implementations of the present disclosure, a non-transient Computer-Readable Medium (CRM) stores instructions for performing the method described in FIG. 13 based on execution by at least one processor.
[0270] FIG. 14 illustrates an example of a method performed by a base station applicable to an implementation of the present disclosure.
[0271] In step S1410, the method includes the step of receiving from the UE information related to one or more alternative actions applicable to the degraded action. The performance prediction values of the one or more alternative actions are derived based on an evaluation model.
[0272] In step S1410, the method includes the step of transmitting configuration information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions to the UE. Based on the configuration information, the optimal alternative action is executed.
[0273] Additionally, the method described in Fig. 14 from the perspective of a base station can be performed by the second wireless device (200) shown in Fig. 2.
[0274] The base station includes at least one transceiver, at least one processor, and at least one memory that stores instructions for the UE to perform the method described in FIG. 13 based on being connected to the at least one processor to be operable and executed by the at least one processor.
[0275] More specifically, the base station receives information from the UE regarding one or more alternative actions applicable to the degraded action. Performance prediction values of the one or more alternative actions are derived based on an evaluation model.
[0276] The base station transmits configuration information related to the optimal alternative action selected based on the performance prediction values of the one or more alternative actions to the UE. The optimal alternative action is executed based on the configuration information.
[0277] Various implementations of the present disclosure are described below.
[0278] The implementation of the present disclosure described below in FIGS. 15 and / or 16 illustrates an example of a signaling procedure between a UE and a network. In FIGS. 15 and / or 16, the UE and the network are merely examples and may be replaced with various devices disclosed in the present disclosure. Additionally, some steps disclosed in FIGS. 15 and / or 16 may be omitted without loss of generality depending on the situation and / or setting, etc.
[0279] FIG. 15 illustrates an example of a procedure to which an implementation of the present disclosure is applied.
[0280] The example in Fig. 15 corresponds to an example of monitoring on the network side based on a model that predicts the performance of an alternative operation. That is, Fig. 15 corresponds to a method of deriving a predicted value of the performance of an alternative operation by executing a model that predicts the performance of an alternative operation on the network side.
[0281] In step S1510, performance degradation of the UE-side AI / ML model is determined.
[0282] The determination of performance degradation of the main model can be made by the entity that calculated the metric or the entity that received the metric. For example, the UE can determine the performance degradation of the UE-side AI / ML model.
[0283] In step S1520, the UE reports information about one or more applicable alternative actions to the network.
[0284] For example, a set of alternate IDs may be defined, and based on this, applicable alternate actions between the UE and the network may be classified. That is, information about one or more alternate actions may be alternate IDs corresponding to each alternate action. Information related to alternate IDs may be transmitted via explicit signaling (e.g., bitmaps). For example, alternate IDs corresponding to the second and fourth candidate models and the first and third legacy methods may be transmitted to the network.
[0285] In step S1530, the UE receives configuration information related to the evaluation model of the alternative action from the network.
[0286] For example, configuration information related to an evaluation model may include at least one of i) the type / category of an evaluation model currently available for operation in the network and / or configuration information required for operating the evaluation model.
[0287] In step S1540, the UE may report additional information to the network for selecting the type of alternative action and / or evaluation model.
[0288] For example, additional information for selecting alternative actions and / or types of evaluation models may include at least one of i) conditions / criterions / thresholds related to the selection of alternative actions or types / categories of preferred evaluation models (e.g., preferred performance ID or criteria).
[0289] Additional information for selecting alternative actions and / or the type of evaluation model may be reported by the UE to the network and / or configured by the network to the UE. That is, in the case of UE-preferred alternative actions, the UE may report additional information to the network as in step S1540. In the case of network-preferred alternative actions, the additional information may be configured by including it in the configuration information related to the evaluation model received in step S1530.
[0290] In step S1550, the network runs a set of evaluation models for the selected alternative actions and derives a prediction value.
[0291] The network can synthesize the information received in the previous step to determine a set of finally viable evaluation models and then execute them. For example, the network can execute a model that predicts the throughput of the second candidate model and a model that predicts the throughput of the first legacy method.
[0292] The network can determine the optimal LCM operation based on the result of the evaluation model. For example, conditions for each LCM operation can be defined according to at least one of the following.
[0293] - Model Switching Condition: The predicted performance of a specific candidate model is the best
[0294] - Fallback condition: The predicted performance of a specific legacy method is the best
[0295] - Model reactivation condition: The principal model's predicted performance is the best (e.g., when the network has an evaluation model of the principal model)
[0296] The network can reactivate the main model when the reliability of the evaluation model decreases through continuous monitoring of the evaluation model. For example, if the difference between the output of a specific evaluation model and the output of an actual alternative action is greater than a threshold, the network determines that the reliability of the evaluation model has decreased and, accordingly, can reactivate the main model or consider other alternative actions.
[0297] In step S1560, the network transmits information related to the determined optimal alternative action to the UE.
[0298] In step S1570, the UE executes the optimal alternative action.
[0299] If it is determined that the performance of the optimal alternative operation has deteriorated, it may be re-executed starting from step S1520.
[0300] FIG. 16 illustrates another example of a procedure to which an implementation of the present disclosure is applied.
[0301] The example in Fig. 16 corresponds to an example where hybrid monitoring is performed based on a model that predicts the performance of an alternative action. That is, Fig. 16 corresponds to a method in which the network transmits a model that predicts the performance of an alternative action to the UE, and the UE executes it to derive a predicted value of the performance of the alternative action.
[0302] In step S1610, performance degradation of the UE-side AI / ML model is determined.
[0303] The determination of performance degradation of the main model can be made by the entity that calculated the metric or the entity that received the metric. For example, the UE can determine the performance degradation of the UE-side AI / ML model.
[0304] In step S1620, the UE reports information about one or more applicable alternative actions to the network.
[0305] For example, a set of alternate IDs may be defined, and based on this, applicable alternate actions between the UE and the network may be classified. That is, information about one or more alternate actions may be alternate IDs corresponding to each alternate action. Information related to alternate IDs may be transmitted via explicit signaling (e.g., bitmaps). For example, alternate IDs corresponding to the second and fourth candidate models and the first and third legacy methods may be transmitted to the network.
[0306] In step S1630, the UE receives configuration information related to the evaluation model of the alternative action from the network.
[0307] For example, configuration information related to an evaluation model may include at least one of i) the type / category of an evaluation model currently available for operation in the network and / or configuration information required for operating the evaluation model.
[0308] In step S1640, the UE may report additional information to the network for selecting the type of alternative action and / or evaluation model.
[0309] For example, additional information for selecting alternative actions and / or types of evaluation models may include at least one of i) conditions / criterions / thresholds related to the selection of alternative actions or types / categories of preferred evaluation models (e.g., preferred performance ID or criteria).
[0310] Additional information for selecting the type of alternative action and / or evaluation model may be reported by the UE to the network and / or set by the network to the UE. That is, in the case of a UE-preferred alternative action, the UE may report additional information to the network as in step S1640. In the case of a network-preferred alternative action, the additional information may be set by including it in the configuration information related to the evaluation model received in step S1630.
[0311] In step S1650, the network transmits a set of evaluation models of the selected alternative actions to the UE.
[0312] The network can synthesize the information received in the previous step to determine a set of finally viable evaluation models and then transmit them to the UE. For example, the network can transmit information related to a model predicting the throughput of the second candidate model and a model predicting the throughput of the first legacy method to the UE.
[0313] In step S1660, the UE executes a set of evaluation models for alternative behaviors.
[0314] In step S1670, the UE reports the result values of the evaluation model set to the network.
[0315] The network can determine the optimal LCM operation based on the result of the evaluation model. For example, conditions for each LCM operation can be defined according to at least one of the following.
[0316] - Model Switching Condition: The predicted performance of a specific candidate model is the best
[0317] - Fallback condition: The predicted performance of a specific legacy method is the best
[0318] - Model reactivation condition: The principal model's predicted performance is the best (e.g., when the network has an evaluation model of the principal model)
[0319] The network can reactivate the main model when the reliability of the evaluation model decreases through continuous monitoring of the evaluation model. For example, if the difference between the output of a specific evaluation model and the output of an actual alternative action is greater than a threshold, the network determines that the reliability of the evaluation model has decreased and, accordingly, can reactivate the main model or consider other alternative actions.
[0320] In step S1680, the network transmits information related to the determined optimal alternative action to the UE.
[0321] In step S1690, the UE executes the optimal alternative action.
[0322] If it is determined that the performance of the optimal alternative operation has deteriorated, it may be re-executed starting from step S1620.
[0323] The present disclosure may have various effects.
[0324] For example, if performance degradation of the main model is detected, an intelligent action decision system differentiated from conventional monitoring methods can be provided by comparing and / or evaluating the predictive performance of all conceivable alternative actions in real time.
[0325] For example, when determining an alternative action, one or more types and / or categories of alternative actions that are operable or preferred by the network side or the terminal may be considered. Through this, an optimal alternative action that matches the capabilities and preferences of the network environment and / or the terminal can be selected, and more flexible and adaptive model management is possible compared to conventional methods that rely on uniform criteria.
[0326] For example, in terms of fallback, by setting multiple legacy actions as a candidate pool and comparing their respective prediction performances, it is possible to support a transition to a legacy action optimized for the current operating environment. Consequently, unnecessary periods of performance degradation can be minimized, and service continuity can be improved.
[0327] For example, in terms of model switching, it is possible to compare and / or evaluate the performance of multiple candidate AI / ML models in real time. This supports dynamic switching to the model best suited to the current channel environment and terminal conditions, and can improve overall system performance and robustness compared to conventional methods that rely on a single model.
[0328] For example, even if the performance of the optimized alternative action itself degrades, it is possible to search for and determine the next-best alternative action by recursively applying the monitoring process. Based on this recursive monitoring structure, the system can continuously maintain an optimal operating state even in situations of multi-stage performance degradation.
[0329] The effects obtainable through the specific examples of the present disclosure are not limited to those listed above. For example, there may be various technical effects that a person having ordinary skill in the related art can understand or derive from the present disclosure. Accordingly, the specific effects of the present disclosure are not limited to those explicitly described in the present disclosure, but may include various effects that can be understood or derived from the technical features of the present disclosure.
[0330] The claims described in this disclosure may be combined in various ways. For example, the technical features of the method claims of this disclosure may be combined to be implemented as a device, and the technical features of the device claims of this disclosure may be combined to be implemented as a method. Additionally, the technical features of the method claims of this disclosure and the technical features of the device claims of this disclosure may be combined to be implemented as a device, and the technical features of the method claims of this disclosure and the technical features of the device claims of this disclosure may be combined to be implemented as a method. Other implementations are within the scope of the following claims.
Claims
1. In a method performed by UE (User Equipment), A step for determining performance degradation of operation; A step of reporting to a network information related to one or more alternative actions applicable to the above action, Performance prediction values of one or more of the above alternative actions are derived based on an evaluation model; A step of receiving configuration information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions from the network; and A method comprising the step of executing the optimal alternative action based on the above setting information.
2. In Paragraph 1, A method further comprising the step of receiving configuration information related to one or more alternative operations from the network.
3. In Paragraph 2, A method comprising at least one of the configuration information related to one or more alternative operations, i) the type and / or category of an evaluation model currently operable by the network, or ii) configuration information required for the operation of the evaluation model.
4. In Paragraph 1, The input of the above evaluation model is identical to the input of the model related to the above operation, and A method in which the output of the above evaluation model consists of performance prediction values of one or more alternative actions.
5. In Paragraph 1, The above performance prediction value is a method derived from the above network.
6. In Paragraph 1, The above performance prediction value is a method derived from the above UE.
7. In Paragraph 6, A method further comprising the step of receiving configuration information related to the above evaluation model from the above network.
8. In Paragraph 6, A method further comprising the step of reporting the above-derived performance prediction value to the above-described network.
9. In Paragraph 1, The above optimal alternative operation is a method comprising at least one of operation switching, fallback, or reactivation of the operation.
10. In Paragraph 1, A method comprising setting information related to the optimal alternative operation, wherein the setting information includes at least one of a condition for switching the operation, a condition for falling back, or a condition for reactivating the operation.
11. In Paragraph 1, A method in which one or more of the above alternative actions are defined based on an alternative ID (identifier) set.
12. In Paragraph 1, A method further comprising the step of reporting information for selecting one or more alternative actions or types of evaluation models.
13. In Paragraph 1, The above operation is a method based on AI (Artificial Intelligence) / ML (Machine Learning) on the UE side.
14. Regarding UE (User Equipment), At least one transmitter / receiver; At least one processor; and It includes at least one memory that can be operably connected to the at least one processor and stores instructions that cause the UE to perform an operation based on execution by the at least one processor. The above operation is: A step for determining performance degradation of operation; A step of reporting to a network information related to one or more alternative actions applicable to the above action, Performance prediction values of one or more of the above alternative actions are derived based on an evaluation model; A step of receiving configuration information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions from the network; and A step of executing the optimal alternative operation based on the above setting information; UE including 15. In the device, At least one processor integrated with the UE (User Equipment); and It includes at least one memory storing processor-executable instructions configured to enable the above at least one processor to perform an operation, and The above operation is: A step for determining performance degradation of operation; A step of generating information related to one or more alternative actions applicable to the above action, Performance prediction values of one or more of the above alternative actions are derived based on an evaluation model; A step of obtaining setting information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions; and A step of executing the optimal alternative operation based on the above setting information; A device including 16. In a non-transitory computer readable medium (CRM) storing instructions that perform operations based on execution by at least one processor, The above operation is: A step for determining performance degradation of operation; A step of reporting to a network information related to one or more alternative actions applicable to the above action, Performance prediction values of one or more of the above alternative actions are derived based on an evaluation model; A step of receiving configuration information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions from the network; and A step of executing the optimal alternative operation based on the above setting information; Non-transient CRM including 17. In a method performed by a base station, A step of receiving information from the UE (User Equipment) regarding one or more alternative actions applicable to a degraded action, Performance prediction values of one or more of the above alternative actions are derived based on an evaluation model; and The method includes the step of transmitting configuration information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions to the UE. A method for executing the optimal alternative action based on the above setting information.
18. Regarding base stations, At least one transmitter / receiver; At least one processor; and It includes at least one memory that can be operably connected to the at least one processor and stores instructions that cause the UE to perform an operation based on execution by the at least one processor. The above operation is: A step of receiving information from the UE (User Equipment) regarding one or more alternative actions applicable to a degraded action, Performance prediction values of one or more of the above alternative actions are derived based on an evaluation model; and A step of transmitting configuration information related to an optimal alternative action selected based on performance prediction values of one or more alternative actions to the UE; Includes, A base station in which the optimal alternative operation is executed based on the above setting information.