Beam management framework with machine learning and model transfer

Abstract

Methods and apparatus are provided for beam management with machine learning. A user equipment (UE) receives, from a wireless network, one or more physical downlink shared channels (PDSCHs) including data of a neural network (NN) model for beam management. The UE verifies the integrity of the NN model received from the wireless network and determines UE capabilities that support the NN model received from the wireless network. In response to verifying the integrity and determining the UE capability to support the NN model, the UE transmits an NN model transfer completion confirmation to the wireless network.

Description

Technical Field

[0001] The present application relates generally to wireless communication systems, including beam management. Background Art

[0002] Wireless mobile communication technology uses various standards and protocols to transmit data between base stations and wireless communication devices. Wireless communication system standards and protocols may include, for example, the 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) (e.g., 4G), 3GPP New Radio (NR) (e.g., 5G), and the IEEE 802.11 standard for wireless local area networks (WLANs) (commonly referred to within industry organizations as WLANs). ).

[0003] As envisioned by 3GPP, different wireless communication system standards and protocols may use various radio access networks (RANs) to communicate between base stations of the RAN (which may also sometimes be referred to as RAN nodes, network nodes, or simply nodes) and wireless communication devices referred to as user equipment (UE). 3GPP RANs may include, for example, Global System for Mobile Communications (GSM), Enhanced Data Rates for GSM Evolution (EDGE) RAN (GERAN), Universal Terrestrial Radio Access Network (UTRAN), Evolved Universal Terrestrial Radio Access Network (E-UTRAN), and / or Next Generation Radio Access Network (NG-RAN).

[0004] Each RAN may use one or more radio access technologies (RATs) to perform communications between base stations and UEs. For example, GERAN implements GSM and / or EDGE RAT, UTRAN implements Universal Mobile Telecommunications System (UMTS) RAT or other 3GPP RAT, E-UTRAN implements LTE RAT (sometimes referred to as LTE), and NG-RAN implements NR RAT (sometimes referred to herein as 5G RAT, 5G NR RAT, or NR). In some deployments, E-UTRAN may also implement NR RAT. In some deployments, NG-RAN may also implement LTE RAT.

[0005] The base stations used by the RAN may correspond to the RAN. An example of an E-UTRAN base station is an Evolved Universal Terrestrial Radio Access Network (E-UTRAN) Node B (also commonly denoted as an evolved Node B, enhanced Node B, eNodeB, or eNB). An example of an NG-RAN base station is a Next Generation Node B (sometimes also referred to as a g-Node B or gNB).

[0006] The RAN provides communication services together with external entities through its connection with the Core Network (CN). For example, E-UTRAN may utilize the Evolved Packet Core (EPC) and NG-RAN may utilize the 5G Core Network (5GC). BRIEF DESCRIPTION OF THE DRAWINGS

[0007] To easily identify the discussion of any particular element or action, the most significant digit(s) in a reference number refers to the drawing number that first introduces the element.

[0008] Figure 1 An encoder and a decoder in CSI feedback operation according to certain embodiments are illustrated.

[0009] Figure 2 is a signaling diagram illustrating UE-side model training and UE-side inference according to certain specific implementations.

[0010] Figure 3 is a signaling diagram illustrating NW-side model training and NW-side inference according to certain specific implementations.

[0011] Figure 4 is a signaling diagram illustrating NW-side model training and UE-side inference according to certain specific implementations.

[0012] Figure 5A and Figure 5B is a signaling diagram illustrating model activation timing according to certain embodiments.

[0013] Figure 6 is a flow chart of a method for a UE to communicate in a wireless network according to one embodiment.

[0014] Figure 7 is a flowchart of a method for a base station to configure beam management in a wireless network according to one embodiment.

[0015] Figure 8 A timeline for starting downloading a NN model to a UE in anticipation of a handover according to one embodiment is illustrated.

[0016] Fig. 9 is a signaling diagram illustrating the transfer and activation of a NN model during handover according to one embodiment.

[0017] Fig.10 is a signaling diagram illustrating the transfer and activation of a NN model during handover according to one embodiment.

[0018] Fig.11 is a flow chart of a method for a UE to communicate in a wireless network according to one embodiment.

[0019] Fig.12 is a flow chart of a method for a base station to communicate with a UE in a wireless network according to one embodiment.

[0020] Fig.13An example architecture of a wireless communication system according to embodiments disclosed herein is illustrated.

[0021] Fig.14 A system for performing signaling between a wireless device and a network device according to embodiments disclosed herein is illustrated. DETAILED DESCRIPTION

[0022] Various embodiments are described with respect to UE. However, reference to UE is provided for illustrative purposes only. The example embodiments may be used with any electronic component that can establish a connection with a network (NW) and is configured with hardware, software and / or firmware for exchanging information and data with the NW. Therefore, UE as described herein is used to represent any appropriate electronic component.

[0023] Downlink channel state information (CSI) (e.g., for frequency division duplex (FDD) operation) may be sent from the UE to the base station via a feedback channel. For example, the base station may use CSI feedback to reduce interference and increase throughput for massive multiple-input multiple-output (MIMO) communications. However, such feedback uses excessive overhead. Vector quantization or codebook-based feedback may be used to reduce feedback overhead. However, the amount of feedback generated by these methods scales linearly with the number of transmit antennas, which may be difficult when using hundreds or thousands of centralized or distributed transmit antennas.

[0024] Artificial intelligence (AI) and / or machine learning (ML) may be used to enhance CSI feedback to reduce overhead, improve accuracy, and / or generate predictions. AI and / or ML may also be used, for example, for beam management (e.g., beam prediction in the time / spatial domain to reduce overhead and latency and improve beam selection accuracy) and / or positioning accuracy enhancement.

[0025] CSI feedback using AI and / or ML can be formulated as a joint optimization of the encoder and decoder. See, for example, Chao-Kai Wen, Wan-Ting Shih, and Shi Jin, “Deep Learning for Massive MIMO CSI Feedback,” IEEE Wireless Communications Letters, Vol. 7, No. 5, October 2018. Since the early paper by Chao-Kai Wen et al., autoencoders and many variants have been considered. Image processing / video processing techniques have been used for CSI compression, which may be a natural choice given the recent wave of ML applications in image processing / video processing. In addition, when formulated in the right domain, CSI feedback has similarities with image / video streams.

[0026] Figure 1The encoder 102 of the UE and the decoder 104 of the base station (e.g., gNB) in the AI-based CSI feedback operation according to certain embodiments are briefly illustrated. As shown, the encoder 102 receives a downlink (DL) channel H and outputs an AI-based CSI feedback. The encoder 102 learns the transformation from the original transformation matrix to the compressed representation (codeword) through training data. The decoder 104 learns the inverse transformation from the codeword to the original channel. Therefore, the decoder 104 can receive the AI-based CSI feedback (codeword) from the encoder 102 and output the reconstructed channel H. The encoder 102 and the decoder 104 can be trained using end-to-end learning (e.g., using an unsupervised learning algorithm). Typically, the normalized mean square error (NMSE) or cosine similarity is an optimization metric. In some designs, the DL channel H can be replaced by a DL precoder. Therefore, the encoder 102 takes the DL precoder as input and generates the AI-based CSI feedback, and the decoder 104 takes the AI-based CSI feedback and reconstructs the DL precoder.

[0027] Various types of neural network (NN) encoders / decoders can be trained for different purposes, with different tradeoffs in complexity, overhead, and performance. Convolutional neural networks (CNNs) can be used, for example, for CSI feedback for frequency domain and spatial domain CSI reference signals (CSI-RS) compression. Other examples include the use of transformers or generative adversarial networks (GANs). Depending on the number and rank of receive antennas, channel feedback or precoding matrix indicator (PMI) feedback can be used. For example, for four receive antennas, rank 1 and rank 2 feedback can potentially use AI NN trained with eigenvectors as input, while rank 3 and rank 4 can potentially use channel state information as input to the trained AI NN. Data preprocessing can be used for the input of the AI ​​model. Preprocessing from the frequency domain to the time domain can be used, and some small paths can be removed before input to the AI ​​NN. The maximum rank indicates the maximum number of layers per UE, which corresponds to the lack of correlation or interference between the antennas of the UE. For example, rank 1 corresponds to a maximum of one spatial layer of the UE, rank 2 corresponds to a maximum of two spatial layers of the UE, rank 3 corresponds to a maximum of three layers of the UE, and rank 4 corresponds to a maximum of four layers of the UE.

[0028] NN-based CNN+RNN (recurrent NN) can be used for CSI-RS compression in the time, frequency and space domains. The input can be a time series with a set of CSI-RS configurations. Preprocessed time series such as frequency domain preprocessing (performing on the time domain and removing small channel taps) and Doppler domain preprocessing can also be applied as AI input. Angular domain preprocessing is also possible, however, angular domain preprocessing may not be efficient in some specific implementations.

[0029] There may be a perceived difference between AI for CSI and IA for beam management (BM). For CSI feedback, the encoder and decoder may be located at different nodes (UE and gNB). For example, see Figure 1 . For beam management, some have proposed that a one-sided model should be used, where the AI ​​model is more or less implementation specific. For example, if the AI ​​model is located at the UE, then training and inference are performed at / by the UE. On the other hand, if the AI ​​model is located at the gNB, then training and inference are performed at / by the gNB. Auxiliary information may be used in some implementations. For example, through internal investigations it was found that the loading with direct Fourier transform (DFT) beams can be unequal. Therefore, the analog beam design may need to be improved or optimized, which may no longer follow the description of utilizing DFT precoding.

[0030] At least two sets of beams can be associated with the NN model. Set A includes beams for which the NN model generates predictions. Set B includes beams that were measured, and the measurements are used as input to the NN model.

[0031] In some systems, for AI / ML based beam management, support is provided for BM-Case 1 and BM-Case 2 for characterization and baseline performance evaluation. BM-Case 1 includes spatial domain DL beam prediction for set A beams based on measurements of set B beams. BM-Case 2 includes temporal DL beam prediction for set A beams based on historical measurements of set B beams. For BM-Case 1 and BM-Case 2, the beams in set A and set B can be in the same frequency range.

[0032] For sub-use case BM-Case 1, two alternatives (Alternative 1 and Alternative 2) can be considered. Alternative 1 is AI / ML inference at the NW side, and Alternative 2 is AI / ML inference at the UE side. Similarly, for sub-use case BM-Case 2, Alternative 1 and Alternative 2 can be considered, where for Alternative 1, AI / ML inference is performed at the NW side, and for Alternative 2, AI / ML inference is performed at the UE side. In certain implementations, with respect to sub-use case BM-Case 2, measurement results of the K (K>=1) most recent measurement instances may be used for AI / ML model input (the value of K depends on the specific implementation). In certain implementations, with respect to sub-use case BM-Case 2, the AI / ML model output may include F predictions for F future time instances, where each prediction is for each time instance (at least F=1, where other values ​​of F depend on the specific implementation). As used herein, AI models, ML models, and / or NN models may be used interchangeably.

[0033] Figure 2is a signaling diagram illustrating UE-side model training and UE-side inference according to certain specific implementations. The example shows a training step (T step) and an inference step (I step). However, those skilled in the art will recognize from the disclosure herein that these steps may be performed in a different order than shown, and there may not be a strict time order requirement between the T step and the I step.

[0034] In training step T-0, UE 204 sends AI capability signaling to network 206. In training step T-1, network 206 responds by sending configuration and reference signal transmissions to UE 204. In training step T-2, based on the configuration and measurements of the reference signals, UE 204 generates and provides training data for AI / ML model training 202. In training step T-3, UE or UE-side server performs training of the NN model. In training step T-4, UE or UE-side server loads or updates the trained NN model into the NN engine of the UE.

[0035] In the inference step I-1, the network 206 sends a reference signal transmission for beam management of one or more control beams and / or data beams (including at least set B). In the inference step I-2, the UE 204 performs inference of the best beam at the UE using the trained NN model. In the inference step I-3, the UE 204 sends a beam report to the network 206 to recommend the best beam or a good beam set. In the inference step I-4, the network 206 sends a beam indication to update the control beam / data beam.

[0036] In this example, AI / ML model training 202 and inference are performed at the UE 204 or UE-side server. Therefore, the simulated beam design information may have been embedded in the training data from T-1. Therefore, no additional auxiliary information about the simulated beam may be required. However, the simulated beam design may be different for different infrastructure vendors, or even different types of modules from the same infrastructure vendor, or different field operations, administration, and maintenance (OAM) configurations for the same type of modules. Therefore, the trained model may be site-specific. In addition, considering that different modules can be used for different frequency bands at the same site under the same operator, the trained model can be frequency band specific. In existing systems, it is unclear how the UE-side server can implement such training.

[0037] For example, if the trained model is site-specific and band-specific, frequent model updates or model switching may be required for UE mobility. In the case of "model update", the storage of the AI ​​model at the UE may be limited, so when the UE moves to a new cell, the new AI model (with new weights) may need to be loaded from the UE-side server in real time. In the case of "model switching", the storage of the AI ​​model at the UE is large enough, so when the UE moves to a new cell, the new AI model (with new weights) stored at the UE is already turned on. Therefore, it may be necessary to load the new AI model from the UE-side server into the UE, but this may not be performed in real time. Compared with the problem of training site-specific / band-specific models, the logistics of model transfer may not be a problem.

[0038] Figure 3 It is a signaling diagram illustrating UW-side model training and UW-side inference according to certain specific implementations. As described above, those skilled in the art will recognize from the disclosure herein that these steps may be performed in a different order than shown, and there may not be a strict time order requirement between the T step and the I step.

[0039] In training step T-0, UE 302 sends AI capability signaling to network 304. In training step T-1, network 304 responds by sending configuration and reference signal transmissions to UE 302. In training step T-2, based on the configuration and measurements of the reference signals, UE 302 generates and sends training data to network 304, which provides the training data to AI / ML model training 306 at NW or NW-side server. In training step T-3, NW or NW-side server performs training of NN model. In training step T-4, NW or NW-side server loads or updates the trained NN model into the NN engine of network 304.

[0040] In the inference step I-1, the network 304 sends a reference signal transmission for beam management of one or more control beams and / or data beams (including at least set B). In the inference step I-2, the UE 302 sends a beam report about the set B beams and optionally other beams to the network 304. In the inference step I-3, the network 304 performs inference using the trained NN model to infer the transmit beam to be sent to the UE 302. In the inference step I-4, the network 304 sends a beam indication to the UE 302 to update the control beam / data beam.

[0041] In this example, with NW-side model training and NW-side inference, the simulated beam design information may have been embedded in the training data. Therefore, the illustrated beam management procedure may be useful, and the enhancement may be limited to T2 (e.g., increasing the number of reported beams). However, Figure 3 The example shown in may increase feedback overhead. For example, before inferring step I-4, if the network 304 is not sure about the inferred transmit beam, the network 304 may transmit multiple candidate beams from set A to the UE 302, and the UE 302 may report the corresponding reference signal received power (RSRP) or the best transmit beam to the network 304. Similarly, for Figure 2 The same need for increased overhead may occur for the UE side example shown in .

[0042] Figure 4 It is a signaling diagram illustrating NW-side model training and UE-side inference according to certain specific implementations. As described above, those skilled in the art will recognize from the disclosure herein that these steps may be performed in a different order than shown, and there may not be a strict time order requirement between the T step and the I step.

[0043] In training step T-0, UE 402 sends AI capability signaling to network 404. In training step T-1, network 404 responds by sending configuration and reference signal transmission to UE 402. In training step T-2, based on the configuration and measurement of the reference signal, UE 402 generates feedback of the training data and sends it to network 404, which provides the feedback to AI / ML model training 406 at NW or NW-side server. In training step T-3, NW or NW-side server performs training of the NN model. In training step T-4, NW or NW-side server sends the trained NN model to UE 402 (e.g., in a direct link from NW-side server to UE) to load or update the trained NN model into the NN engine of UE 402.

[0044] In the inference step I-1, the network 404 sends a reference signal transmission for beam management of one or more control beams and / or data beams (including at least set B). In the inference step I-2, the UE 402 performs inference of the best beam at the UE 402 using the trained NN model. In the inference step I-3, the UE 402 sends a beam report to the network 404 to recommend the best beam or a good beam set. In the inference step I-4, the network 404 sends a beam indication to the UE 402 to update the control beam / data beam.

[0045] In this example, with NW-side model training and UE-side inference, simulated beam design information may have been embedded in the training data from T-1. Therefore, the illustrated beam management procedure may be useful, and enhancements may be limited to T2 (e.g., increasing the number of reported beams). However, Figure 4The example shown in may increase feedback overhead. For example, before inferring step I-4, if the network 304 is not sure about the inferred transmit beam, the network 404 may transmit multiple candidate beams from set A to the UE 402, and the UE 402 may report the corresponding reference signal received power (RSRP) or the best transmit beam to the network 404.

[0046] Conceptually, it is possible to infer the best transmit beam assuming there are no abrupt changes in the two-dimensional (2D) plane or in the data itself. In some implementations, 2D data may facilitate better NN models. In an example use case, a shallow network or NN with four layers may produce good results compared to AI for CSI where transformers have been considered. In some implementations, the AI ​​model size may not be too large and frequent model transfers may be feasible.

[0047] Therefore, in certain embodiments disclosed herein, the network trains an AI model for a single cell or multiple cells. In one embodiment, the AI ​​model for a single cell is associated with a physical cell identifier (PCI). In another embodiment, the AI ​​model for a single cell is associated with multiple transmit receive points (mTRPs). Additionally or in other embodiments, the AI ​​models for multiple cells may be associated with cells with different geographic coverage. Additionally or in other embodiments, the AI ​​models for multiple cells may be associated with cells in the same frequency band.

[0048] In certain embodiments, when the UE enters connected mode, the UE may receive an AI model for a single cell or multiple cells from the network. In one embodiment, the UE uses the AI ​​model to derive RSRP or beam index. Additionally or in other embodiments, the UE may be configured by the network to monitor the performance of the AI ​​model.

[0049] In some cellular systems, in both the case of NW-side training / inference and UE-side training / inference, model transfer never or rarely occurs, even if NW-side training / UE-side inference or UE-side training / NW-side inference is considered not serious when model generalization. Now, with the understanding of analog beam design and the expected difficulty of generalization, frequent model transfer can be triggered. Therefore, the timing of model activation (i.e., model activation delay) needs to be considered.

[0050] Figure 5A and Figure 5Bis a signaling diagram illustrating model activation timing according to certain embodiments. These examples use NW-side AI / ML model training 506 (e.g., at a base station or NW-side server), and in training step T-4, a message is sent from the NW or NW-side server (e.g., to the base station) to load or update the trained NN model. Although not shown, certain embodiments may also include, for example Figure 3 or Figure 4 The training steps T-0, T-1, T2 and / or T3 shown in .

[0051] In response to the training step T-4, the network 504 (e.g., a base station) transmits the trained NN model to the UE 502 in one or more physical downlink shared channels (PDSCHs) 508. The one or more PDSCHs 508 are exemplified by PDSCH-1 to PDSCH-N carrying data corresponding to respective portions of the NN model, wherein the UE 502 confirms successful reception of each portion by sending a hybrid automatic repeat request (HARQ) acknowledgement (ACK) in response to each PDSCH.

[0052] After receiving the trained NN model, in box 510, UE 502 performs integrity verification of the received NN model and checks the UE's ability to support the NN model. If UE 502 is able to verify the integrity of the NN model and determines that the UE supports the NN model, UE 502 sends an NN model transfer completion confirmation message 512 to the NW-side server. Message 512 can be transparent to the base station (the base station of network 504 only forwards message 512 to the NW-side server). On the other hand, if UE 502 determines at box 510 that the integrity of the NN model cannot be verified and / or UE 502 cannot support the NN model (for example, the version of the NN model is incorrect), message 512 includes a negative confirmation, so that the NW or the NW-side server can try different configurations / versions or can try to train different NN models. At box 514, UE 502 performs NN model compilation for the UE's platform and loads the compiled NN model into the NN engine at UE 502.

[0053] exist Figure 5A In , after the compiled NN model is loaded into the NN engine, the UE 502 sends a message 516 to the network 504 that the NN model is ready for use. Thus, the network 504 knows when it can start transmission of reference signals for beam management (including at least set B). See, e.g. Figure 2 The inference steps I-1, I-2, I-3 and I-4 in .

[0054] exist Figure 5BIn the example, instead of sending the NN model transfer completion confirmation message 512 to the NW side server, the UE 502 sends the message 512 to the network 504 (i.e., to the base station). In response to receiving the NN model transfer completion confirmation message 512 (shown as "Tick"), the base station starts a timer corresponding to the time gap 518 based on the expected delay for compiling the NN model and loading it into the NN engine at the UE 502. The time gap 518 can be derived, for example, from the UE capabilities reported by the UE 502 to the network 504. When the timer expires (shown as "Tock"), the base station determines that the NN model is ready for use. Therefore, the network 504 does not need to Figure 5A The message 516 shown in FIG. 1 is used to know when it can start transmission of reference signals for beam management (including at least set B), which reduces signaling overhead.

[0055] Figure 6 is a flow chart of a method 600 for a UE to communicate in a wireless network according to one embodiment. In box 602, the method 600 includes receiving one or more PDSCHs including data of an NN model for beam management at the UE from the wireless network. In box 604, the method 600 includes verifying the integrity of the NN model received from the wireless network at the UE. In box 606, the method 600 includes determining, at the UE, the UE's capability to support the NN model received from the wireless network. In box 608, in response to verifying the integrity and determining the UE's ability to support the NN model, the method 600 includes transmitting an NN model transfer completion confirmation to the wireless network.

[0056] In certain embodiments, method 600 also includes: generating a compiled NN model by compiling data for the NN model used by the UE; and loading the compiled NN model into the NN engine of the UE.

[0057] In certain embodiments of method 600, transmitting the NN model transfer completion confirmation includes transmitting the NN model transfer completion confirmation from the UE to a server of the wireless network through the base station.

[0058] In some embodiments, method 600 also includes: after loading the compiled NN model into the NN engine of the UE, signaling from the UE to the base station that the NN model is ready for use at the UE.

[0059] In certain embodiments of method 600, transmitting the NN model transfer completion confirmation includes transmitting the NN model transfer completion confirmation to a base station of the wireless network.

[0060] Certain embodiments of method 600 also include: generating a compiled NN model; and loading the compiled NN model into the NN engine of the UE within a time slot signaled to the base station in the UE capability report.

[0061] In certain embodiments of method 600, the NN model is trained by the wireless network for a single cell, and the NN model is associated with a physical cell identifier (PCI).

[0062] In certain embodiments of method 600, the NN model is trained by the wireless network for a single cell, and the NN model is associated with multiple transmit receive points (mTRPs) in the single cell of the wireless network.

[0063] In certain embodiments of method 600, the NN model is trained by the wireless network for multiple cells having different geographic coverage.

[0064] In certain embodiments of method 600, the NN model is trained by the wireless network for multiple cells associated with the same frequency band.

[0065] In certain embodiments, method 600 also includes deriving, at the UE, at least one of a reference signal received power (RSRP) and one or more beam indices using the compiled NN model.

[0066] In some embodiments, method 600 also includes receiving a configuration for monitoring performance of the NN model from the wireless network.

[0067] Figure 7 is a flow chart of a method 700 for a base station to configure beam management in a wireless network according to one embodiment. In box 702, the method 700 includes receiving, at the base station, an indication from a server in the wireless network to load or update a NN model for beam management at a UE. In box 704, in response to the indication from the server, the method 700 includes sending one or more PDSCHs including data for the NN model from the base station to the UE. In box 706, the method 700 includes determining that the UE is ready to use the NN model. In box 708, in response to determining that the UE is ready to use the NN model, the method 700 includes transmitting one or more reference signals (RS) to the UE.

[0068] In certain embodiments, method 700 also includes: forwarding a NN model transfer completion confirmation from the UE to a server in the wireless network in a transparent manner; and receiving a signal from the UE at the base station indicating that the UE is ready to use the NN model.

[0069] In certain embodiments, method 700 also includes: receiving a UE capability report indicating a time gap value from the UE at a base station; receiving an NN model transfer completion confirmation from the UE at the base station; in response to receiving the NN model transfer completion confirmation, starting a timer corresponding to the time gap value; and when the timer expires, determining that the UE is ready to use the NN model.

[0070] In certain embodiments of method 700, the NN model is trained by a server in the wireless network for a single cell, and the NN model is associated with a physical cell identifier (PCI).

[0071] In certain embodiments of method 700, the NN model is trained by a server in the wireless network for a single cell, and the NN model is associated with multiple transmit receive points (mTRPs) in the single cell of the wireless network.

[0072] In certain embodiments of method 700, the NN model is trained by a server in a wireless network for multiple cells with different geographic coverage.

[0073] In certain embodiments of method 700, the NN model is trained by a server in the wireless network for multiple cells associated with the same frequency band.

[0074] In some embodiments, method 700 also includes transmitting, from the base station to the UE, a configuration for monitoring performance of the NN model.

[0075] Certain embodiments provide for NN model transfer during handover. Conditional handover is used to provide a set of configurations for a target cell to the UE before the handover actually occurs. Similar to conditional handover, embodiments disclosed herein may use the network's prediction to identify a target cell for handover. In response to identifying the target cell, the network begins downloading the NN model configured for the target cell to the UE.

[0076] For example, Figure 8 A timeline for starting to download NN models to a UE in anticipation of a handover according to one embodiment is illustrated. At a first time T1, the UE uses an NN engine with a first activated NN model (NN model-1). At a second time T2, based on the measurement values ​​reported from the UE, the network predicts that the UE will move from a source cell to a target cell. In response, at a third time T3, the network starts downloading a second NN model (NN model-2) to the UE. Therefore, since the second NN model is downloaded to the UE before the handover, the delay in activating the second NN model after the handover is reduced.

[0077] Fig. 9is a signaling diagram illustrating the transfer and activation of NN models during handover according to one embodiment. In this example, the UE 902 may use a first NN model (NN model-1) in a current or source cell, and the network 904 or NW-side server may provide AI / ML model training 906. Various instances of the NN memory and NN engine of the UE 902 are shown for illustrative purposes. In other words, the NN memory is shown as a different time from the NN memory 908a, NN memory 908b, NN memory 908c, NN memory 908d, and NN memory 908e. Similarly, the NN engine is shown as a different time from the NN engine 910a, NN engine 910b, NN engine 910c, NN engine 910d, and NN engine 910e. For example, in one instance, the NN memory 908a stores NN model-X, and the NN engine 910a operates with the activated NN model-1. Certain embodiments use a single NN engine or a single line or processing thread in the NN engine. Other embodiments (e.g., see Fig.10 ) Using multiple NN engines by using multiple processing cores within the UE's processor or using multiple processors at the UE to perform parallel processing.

[0078] At block 912, the network 904 predicts that the UE will move to the target cell and determines that NN Model-X is not suitable for the target cell. In response, the network 904 begins 914 downloading a second NN model (NN Model-2) to the UE 902. For example, the network 904 may transmit one or more PDSCHs carrying NN Model-2 to the UE 902 (see Figure 5A and Figure 5B ). UE 902 receives NN model-2 and stores it in NN memory 908b, and sends a NN model-2 transfer completion confirmation message 916 to the network 904.

[0079] At box 918, the network 904 and the UE 902 perform a handover of the UE 902 from the source cell to the target cell. After the handover, the network 904 sends a signal 920 to the UE 902 to activate NN model-2. As shown, at the first moment after receiving the signal 920, the NN memory 908c still stores NN model-2, and the NN model-1 is still activated in the NN engine 910c. In response to the signal 920, at the second instance, the UE 902 loads the NN model-2 from the NN memory 908d to the NN engine 910d, and the NN engine deactivates the NN model-1. At the third instance, the NN memory 908e is empty, and NN mode 2 is activated in the NN engine 910e. The UE 902 can then send a message 922 to the network 904 to report that the NN model-2 is ready.

[0080] In some embodiments, the UE may report support for M activated NN reference models to the network (e.g., in a UE capability message). For inter-band carrier aggregation (CA) or dual connectivity (DC), the analog beamforming may be different for different bands or cell sites. Therefore, one reference model may be used as a unit to quantify the complexity of the NN model. For example, NN model-1 may be considered as 1X the reference model, model-2 may be considered as 2X the reference model, and so on. Therefore, the downloaded models may be quantized into multiple reference models.

[0081] Fig.10 1002 is a signaling diagram illustrating the transfer and activation of multiple NN engines of NN models during handover according to one embodiment. For example, multiple NN engines may be provided by using multiple processing cores within a processor of UE 1002 or by using multiple processors at UE 1002 to perform parallel processing. Fig.10 In the example of , various instances of the first NN engine (NN engine-1) are shown as NN engine-1 1010a, NN engine-1 1010b, NN engine-1 1010c, NN engine-1 1010d, and NN engine-1 1010e. Similarly, various instances of the second NN engine (NN engine-2) are shown as NN engine-2 1012a, NN engine-2 1012b, NN engine-2 1012c, NN engine-2 1012d, and NN engine-2 1012e. In addition, different instances of NN memory are shown as NN memory 1008a and NN memory 1008b.

[0082] like Fig.10 As shown, UE 1002 may use NN engine-1 1010a with a first NN model (NN model-1) in a current or source cell, at which time NN engine-2 1012a has a deactivated NN model, and NN memory 1008a stores NN model-X. Network 1004 or a NW-side server may provide AI / ML model training 1006. At box 1014, network 1004 predicts that UE 1002 will move to a target cell and determines that NN model-X is not suitable for the target cell. In response, network 1004 begins 1016 downloading a second NN model (NN model-2) to UE 1002. For example, network 1004 may transmit one or more PDSCHs carrying NN model-2 to UE 1002 (see Figure 5A and Figure 5B ). UE 1002 receives NN model-2 and stores it in NN memory 1008b, and sends a NN model-2 transfer completion confirmation message 1018 to the network 1004.

[0083] At box 1020, the network 1004 and UE 1002 perform a handover of UE 1002 from the source cell to the target cell. After the handover, the network 1004 sends a signal 1022 to UE 1002 to activate NN model-2. As shown, at a first moment (T1) after receiving the signal 1022, NN engine-1 1010c still has NN model-1 activated, and NN engine-2 1012c has NN model-2 (not yet activated). The second moment (T2) is shown to illustrate that there may be some delay from deactivating NN model-1 in NN engine-1 1010d and activating NN model-2 in NN engine-2 1012d before NN model-2 is ready for use. However, in some embodiments, T2 can be zero. When NN Model-1 is deactivated in NN Engine-1 1010e and Model-2 is activated in NN Engine-2 1012e, UE 1002 sends a message 1024 to the network 1004 to indicate that NN Model-2 is ready.

[0084] Fig.11 11 is a flowchart of a method 1100 for a UE to communicate in a wireless network according to one embodiment. In box 1102, the method 1100 includes performing beam management in a first cell of the wireless network using a first NN model activated for an NN engine of the UE at the UE. In box 1104, the method 1100 includes downloading from the wireless network a second NN model configured for a second cell predicted for the UE. In box 1106, the method 1100 includes storing the second NN model in a memory of the UE. In box 1108, the method 1100 includes performing a handover of the UE from the first cell to the second cell of the wireless network. In box 1110, the method 1100 includes receiving a signal activating the second NN model from the wireless network at the UE in response to the handover. In box 1112, in response to the signal, the method 1100 includes activating the second NN model of the NN engine for the UE.

[0085] In some embodiments, method 1100 further includes: in response to downloading the second NN model, transmitting a NN model transfer completion confirmation to the wireless network.

[0086] In certain embodiments of method 1100, activating the second NN model in response to a signal from the wireless network includes: deactivating the first NN model in the NN engine; loading the second NN model from a memory of the UE to the NN engine; and activating the second NN model for use by the NN engine of the UE.

[0087] In certain embodiments of method 1100, the UE is configured for parallel processing, wherein the NN engine includes a first NN engine of a first process, and wherein activating a second NN model includes: loading the second NN model from a memory of the UE to a second NN engine of the second process; activating the second NN model used by the second NN engine; and in response to a signal from the wireless network, switching from the first processing to the second processing to perform beam management in a second cell of the wireless network using the second NN model.

[0088] In certain embodiments, method 1100 also includes performing parallel processing using multiple processing cores within a processor of the UE or using multiple processors at the UE. Certain such embodiments also include reporting from the UE to the base station the UE capability of supporting a number M of activated NN reference models, wherein a first NN reference model is used as a unit for quantizing the NN models, and wherein the number M of activated NN reference models is based on at least one of a size and complexity of the first NN reference model. The second NN model may be quantized to the first NN reference model, and the quantization may be with respect to complexity and / or memory storage.

[0089] Fig.12 is a flow chart of a method 1200 for a base station to communicate with a UE in a wireless network according to one embodiment. In box 1202, the method 1200 includes configuring the UE to use a first NN model in a first cell of the wireless network for beam management. In box 1204, based on feedback from the UE, the method 1200 includes predicting that the UE will move from the first cell of the wireless network to a second cell. In box 1206, in response to predicting that the UE will move to the second cell, the method 1200 includes downloading a second NN model to the UE. In box 1208, when the UE switches from the first cell to the second cell, the method 1200 includes sending a signal from the base station to the UE to activate the second NN model.

[0090] In certain embodiments of method 1200, downloading the second NN model to the UE is also responsive to determining that the first NN model or the third NN model stored by the UE is not configured for the second wireless network.

[0091] In some embodiments, method 1200 also includes receiving, at the base station, a signal from the UE indicating that the second NN model is ready for use at the UE.

[0092] In certain embodiments, method 1200 also includes: receiving a UE capability report indicating a time gap value from the UE at a base station; receiving an NN model transfer completion confirmation from the UE at the base station; in response to receiving the NN model transfer completion confirmation, starting a timer corresponding to the time gap value; and when the timer expires, determining that the UE is ready to use the NN model.

[0093] In certain embodiments, method 1200 also includes: receiving, at a base station, a UE capability report from the UE indicating the number M of activated NN reference models supported, wherein a first NN reference model is used as a unit for quantizing the NN model; and selecting one or more additional NN models to be downloaded to the UE based on the number M of activated NN reference models and at least one of the size and complexity of the first reference model, wherein the second NN model and the one or more additional NN models are quantized to the first reference model in terms of complexity and storage.

[0094] Fig.13 An exemplary architecture of a wireless communication system 1300 according to an embodiment disclosed herein is illustrated. The description provided below is for an exemplary wireless communication system 1300 operating in conjunction with the LTE system standard and / or the 5G or NR system standard provided in the 3GPP technical specifications.

[0095] like Fig.13 As shown, the wireless communication system 1300 includes UE 1302 and UE 1304 (although any number of UEs may be used). In this example, UE 1302 and UE 1304 are illustrated as smart phones (e.g., handheld touch screen mobile computing devices capable of connecting to one or more cellular networks), but may also include any mobile or non-mobile computing device configured for wireless communication.

[0096] UE 1302 and UE 1304 may be configured to be communicatively coupled to RAN 1306. In an embodiment, RAN 1306 may be NG-RAN, E-UTRAN, etc. UE 1302 and UE 1304 utilize connections (or channels) (shown as connection 1308 and connection 1310, respectively) with RAN 1306, where each connection (or channel) includes a physical communication interface. RAN 1306 may include one or more base stations (such as base station 1312 and base station 1314) that implement connection 1308 and connection 1310.

[0097] In this example, connection 1308 and connection 1310 are air interfaces that enable such communicative coupling and may conform to the RAT used by RAN 1306, such as, for example, LTE and / or NR.

[0098] In some embodiments, UE 1302 and UE 1304 may also directly exchange communication data via side link interface 1316. UE 1304 is shown as being configured to access an access point (shown as AP 1318) via connection 1320. By way of example, connection 1320 may include a local wireless connection, such as a connection conforming to any IEEE 802.11 protocol, wherein AP 1318 may include In this example, AP 1318 can connect to another network (eg, the Internet) without going through CN 1324.

[0099] In an embodiment, UE 1302 and UE 1304 may be configured to communicate with each other or with base station 1312 and / or base station 1314 over a multi-carrier communication channel using orthogonal frequency division multiplexing (OFDM) communication signals according to various communication techniques, such as but not limited to orthogonal frequency division multiple access (OFDMA) communication techniques (e.g., for downlink communication) or single carrier frequency division multiple access (SC-FDMA) communication techniques (e.g., for uplink and ProSe or sidelink communication), although the scope of the embodiment is not limited in this respect. OFDM signals may include multiple orthogonal subcarriers.

[0100] In some embodiments, all or part of base station 1312 or base station 1314 may be implemented as one or more software entities running on a server computer as part of a virtual network. In addition, or in other embodiments, base station 1312 or base station 1314 may be configured to communicate with each other via interface 1322. In an embodiment where wireless communication system 1300 is an LTE system (e.g., when CN 1324 is an EPC), interface 1322 may be an X2 interface. The X2 interface may be defined between two or more base stations (e.g., two or more eNBs, etc.) connected to the EPC and / or between two eNBs connected to the EPC. In an embodiment where wireless communication system 1300 is an NR system (e.g., when CN 1324 is a 5GC), interface 1322 may be an Xn interface. The Xn interface is defined between two or more base stations (e.g., two or more gNBs, etc.) connected to a 5GC, between a base station 1312 (e.g., a gNB) and an eNB connected to a 5GC, and / or between two eNBs connected to a 5GC (e.g., CN 1324).

[0101] RAN 1306 is shown as being communicatively coupled to CN 1324. CN 1324 may include one or more network elements 1326 configured to provide various data and telecommunication services to customers / subscribers (e.g., UE 1302 and users of UE 1304) connected to CN 1324 via RAN 1306. The components of CN 1324 may be implemented in one physical device or separate physical devices including components for reading and executing instructions from a machine-readable or computer-readable medium (e.g., a non-transitory machine-readable storage medium).

[0102] In an embodiment, CN 1324 may be an EPC, and RAN 1306 may be connected to CN 1324 via an S1 interface 1328. In an embodiment, S1 interface 1328 may be divided into two parts: an S1 user plane (S1-U) interface, which carries traffic data between base station 1312 or base station 1314 and a serving gateway (S-GW); and an S1-MME interface, which is a signaling interface between base station 1312 or base station 1314 and a mobility management entity (MME).

[0103] In an embodiment, CN 1324 may be a 5GC, and RAN 1306 may be connected to CN 1324 via an NG interface 1328. In an embodiment, NG interface 1328 may be divided into two parts: an NG user plane (NG-U) interface, which carries traffic data between base station 1312 or base station 1314 and a user plane function (UPF); and an S1 control plane (NG-C) interface, which is a signaling interface between base station 1312 or base station 1314 and an access and mobility management function (AMF).

[0104] In general, the application server 1330 may be an element that provides applications (e.g., packet-switched data services) that use Internet Protocol (IP) bearer resources with the CN 1324. The application server 1330 may also be configured to support one or more communication services (e.g., VoIP sessions, group communication sessions, etc.) for the UE 1302 and the UE 1304 via the CN 1324. The application server 1330 may communicate with the CN 1324 via the IP communication interface 1332.

[0105] Fig.14 A system 1400 for performing signaling 1434 between a wireless device 1402 and a network device 1418 according to an embodiment disclosed herein is illustrated. The system 1400 may be part of a wireless communication system as described herein. The wireless device 1402 may be, for example, a UE of a wireless communication system. The network device 1418 may be, for example, a base station (e.g., an eNB or a gNB) of a wireless communication system.

[0106] The wireless device 1402 may include one or more processors 1404. The processor 1404 may execute instructions to cause various operations of the wireless device 1402 to be performed, as described herein. The processor 1404 may include one or more baseband processors implemented using, for example, a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a controller, a field programmable gate array (FPGA) device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.

[0107] The wireless device 1402 may include a memory 1406. The memory 1406 may be a non-transitory computer-readable storage medium that stores instructions 1408 (which may include, for example, instructions executed by the processor 1404). The instructions 1408 may also be referred to as program code or a computer program. The memory 1406 may also store data used by the processor 1404 and results computed by the processor.

[0108] The wireless device 1402 may include one or more transceivers 1410, which may include radio frequency (RF) transmitter and / or receiver circuitry that uses an antenna 1412 of the wireless device 1402 to facilitate signaling (e.g., signaling 1434) transmitted and / or received by the wireless device 1402 with other devices (e.g., network device 1418) according to a corresponding RAT.

[0109] The wireless device 1402 may include one or more antennas 1412 (e.g., one, two, four, or more). For embodiments with multiple antennas 1412, the wireless device 1402 may exploit the spatial diversity of such multiple antennas 1412 to transmit and / or receive multiple different data streams on the same time-frequency resources. This behavior may be referred to as, for example, multiple-input multiple-output (MIMO) behavior (referring to multiple antennas used at each of the transmitting device and the receiving device to implement this aspect). MIMO transmissions by the wireless device 1402 may be implemented based on precoding (or digital beamforming) applied at the wireless device 1402, which multiplexes the data streams across the antennas 1412 based on known or assumed channel characteristics, so that each data stream is received at an appropriate signal strength relative to the other streams and at a desired location in the spatial domain (e.g., the location of the receiver associated with the data stream). Certain embodiments may use a single-user MIMO (SU-MIMO) approach (where data streams are all directed to a single receiver) and / or a multi-user MIMO (MU-MIMO) approach (where separate data streams may be directed to separate (different) receivers in different locations in the spatial domain).

[0110] In certain embodiments with multiple antennas, the wireless device 1402 may implement analog beamforming techniques whereby the phases of signals transmitted by the antennas 1412 are relatively adjusted so that the (joint) transmissions of the antennas 1412 can be directed (this is sometimes referred to as beam steering).

[0111] The wireless device 1402 may include one or more interfaces 1414. The interfaces 1414 may be used to provide input to or output from the wireless device 1402. For example, a wireless device 1402 that is a UE may include an interface 1414, such as a microphone, a speaker, a touch screen, buttons, etc., to allow a user of the UE to provide input and / or output to the UE. Other interfaces of such a UE may be composed of transmitters, receivers, and other circuits that allow the UE to communicate with other devices (e.g., in addition to the transceiver 1410 / antenna 1412 described above), and may communicate according to known protocols (e.g., etc.) to perform the operation.

[0112] The wireless device 1402 may include a beam management module 1416. The beam management module 1416 may be implemented by hardware, software, or a combination thereof. For example, the beam management module 1416 may be implemented as a processor, circuit, and / or instructions 1408 stored in the memory 1406 and executed by the processor 1404. In some examples, the beam management module 1416 may be integrated within the processor 1404 and / or the transceiver 1410. For example, the beam management module 1416 may be implemented by a combination of software components (e.g., executed by a DSP or general purpose processor) and hardware components (e.g., logic gates and circuits) within the processor 1404 or the transceiver 1410.

[0113] The beam management module 1416 may be used in various aspects of the present disclosure, for example, Figure 5A , Figure 5B , Figure 6 , Figure 8 , Fig. 9 and / or Fig.11 The beam management module 1416 may include, for example, a NN engine, a NN model training and / or inference module, a timer, or other components discussed herein.

[0114] The network device 1418 may include one or more processors 1420. The processor 1420 may execute instructions to cause various operations of the network device 1418 to be performed, as described herein. The processor 1420 may include one or more baseband processors implemented using, for example, a CPU, a DSP, an ASIC, a controller, an FPGA device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.

[0115] The network device 1418 may include a memory 1422. The memory 1422 may be a non-transitory computer-readable storage medium that stores instructions 1424 (which may include, for example, instructions executed by the processor 1420). The instructions 1424 may also be referred to as program code or a computer program. The memory 1422 may also store data used by the processor 1420 and results calculated by the processor.

[0116] The network device 1418 may include one or more transceivers 1426, which may include RF transmitter and / or receiver circuits that use an antenna 1428 of the network device 1418 to facilitate signaling (e.g., signaling 1434) transmitted and / or received by the network device 1418 with other devices (e.g., wireless device 1402) according to a corresponding RAT.

[0117] The network device 1418 may include one or more antennas 1428 (e.g., one, two, four, or more). In embodiments with multiple antennas 1428, the network device 1418 may perform MIMO, digital beamforming, analog beamforming, beam steering, etc. as described.

[0118] The network device 1418 may include one or more interfaces 1430. The interface 1430 may be used to provide input to or output from the network device 1418. For example, the network device 1418 as a base station may include an interface 1430 composed of a transmitter, a receiver, and other circuits (e.g., in addition to the transceiver 1426 / antenna 1428 described above), which enables the base station to communicate with other equipment in the core network and / or enables the base station to communicate with an external network, computer, database, etc., in order to achieve the purpose of performing operations, management, and maintenance of the base station or other equipment operably connected to the base station.

[0119] The network device 1418 may include a beam management module 1432. The beam management module 1432 may be implemented by hardware, software, or a combination thereof. For example, the beam management module 1432 may be implemented as a processor, circuit, and / or instructions 1424 stored in the memory 1422 and executed by the processor 1420. In some examples, the beam management module 1432 may be integrated within the processor 1420 and / or the transceiver 1426. For example, the beam management module 1432 may be implemented by a combination of software components (e.g., executed by a DSP or general purpose processor) and hardware components (e.g., logic gates and circuits) within the processor 1420 or the transceiver 1426.

[0120] The beam management module 1432 may be used in various aspects of the present disclosure, for example, Figure 5A , Figure 5B , Figure 7 , Figure 8 , Fig. 9 and / or Fig.12 The beam management module 1432 may include, for example, a NN model training and / or inference module, a NW-side server, or other components discussed herein.

[0121] Embodiments contemplated herein include an apparatus comprising means for performing one or more elements of method 600 and / or method 1100. The apparatus may be, for example, an apparatus that is a UE (such as wireless device 1402 as a UE, as described herein).

[0122] Embodiments contemplated herein include one or more non-transitory computer-readable media including instructions that, when executed by one or more processors of an electronic device, cause the electronic device to perform one or more elements of method 600 and / or method 1100. The non-transitory computer-readable medium may be, for example, a memory of a UE (such as memory 1406 of wireless device 1402 as a UE, as described herein).

[0123] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuits for performing one or more elements of method 600 and / or method 1100. The apparatus may be, for example, an apparatus of a UE (such as wireless device 1402 as a UE, as described herein).

[0124] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 600 and / or the described method 1100. The apparatus may be, for example, an apparatus of a UE (such as the wireless device 1402 as a UE, as described herein).

[0125] Embodiments contemplated herein include a signal as described in or associated with one or more elements of method 600 and / or method 1100 .

[0126] Embodiments contemplated herein include a computer program or computer program product including instructions, wherein when the program is executed by a processor, the processor causes the processor to perform one or more elements of method 600 and / or method 1100. The processor may be a processor of a UE (such as processor 1404 of wireless device 1402 as a UE, as described herein). These instructions may be located, for example, in a processor and / or on a memory of a UE (such as memory 1406 of wireless device 1402 as a UE, as described herein).

[0127] Embodiments contemplated herein include an apparatus comprising means for performing one or more elements of method 700 and / or method 1200. The apparatus may be, for example, an apparatus that is a base station (such as network device 1418 that is a base station, as described herein).

[0128] Embodiments contemplated herein include one or more non-transitory computer-readable media including instructions that, when executed by one or more processors of an electronic device, cause the electronic device to perform one or more elements of method 700 and method 1200. For example, the non-transitory computer-readable medium may be a memory of a base station (such as memory 1422 of network device 1418 acting as a base station, as described herein).

[0129] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuits for performing one or more elements of methods 700 and 1200. The apparatus may be, for example, an apparatus that is a base station (such as network device 1418 as a base station, as described herein).

[0130] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of method 700 and / or method 1200. The apparatus may be, for example, an apparatus that is a base station (such as network device 1418 that is a base station, as described herein).

[0131] Embodiments contemplated herein include a signal as described in or associated with one or more elements of method 700 and / or method 1200 .

[0132] Embodiments contemplated herein include a computer program or computer program product including instructions, wherein execution of the program by a processing element causes the processing element to perform one or more elements of method 700 or method 1200. The processor may be a processor of a base station (such as processor 1420 of network device 1418 as a base station, as described herein). These instructions may be located, for example, in a processor and / or on a memory of a base station (such as memory 1422 of network device 1418 as a base station, as described herein).

[0133] For one or more embodiments, at least one of the components set forth in one or more of the foregoing figures may be configured to perform one or more operations, techniques, processes, and / or methods as described herein. For example, a baseband processor as described herein in conjunction with one or more of the foregoing figures may be configured to operate according to one or more of the examples set forth herein. For another example, a circuit system associated with a UE, a base station, a network element, etc. as described above in conjunction with one or more of the foregoing figures may be configured to operate according to one or more of the examples set forth herein.

[0134] Unless otherwise expressly stated, any of the above embodiments may be combined with any other embodiment (or combination of embodiments). The foregoing description of one or more specific implementations provides illustration and description, but is not intended to be exhaustive or to limit the scope of the embodiments to the precise form disclosed. In view of the above teachings, modifications and variations are possible or can be obtained from the practice of various embodiments.

[0135] Embodiments and implementations of the systems and methods described herein may include various operations that may be embodied in machine executable instructions to be executed by a computer system. A computer system may include one or more general or special purpose computers (or other electronic devices). A computer system may include hardware components that include specific logic components for performing operations; or may include a combination of hardware, software, and / or firmware.

[0136] It should be appreciated that the systems described herein include descriptions of specific embodiments. These embodiments may be combined into a single system, partially combined into other systems, separated into multiple systems, or otherwise divided or combined. In addition, it is contemplated that parameters, attributes, aspects, etc. of another embodiment may be used in one embodiment. For clarity, these parameters, attributes, aspects, etc. are described only in one or more embodiments, and it should be appreciated that these parameters, attributes, aspects, etc. may be combined with or substituted for parameters, attributes, aspects, etc. of another embodiment unless expressly stated herein.

[0137] It is understood that the use of personally identifiable information should be subject to privacy policies and practices that are generally recognized to meet or exceed industry or government requirements for maintaining user privacy. Specifically, personally identifiable information data should be managed and processed to minimize the risk of unintentional or unauthorized access or use, and the nature of the authorized use should be clearly stated to users.

[0138] Although the foregoing has been described in considerable detail for the sake of clarity, it will be apparent that certain changes and modifications may be made without departing from the principles of the invention. It should be noted that there are many alternative ways to implement both the processes and the apparatus described herein. Therefore, the embodiments of the present invention should be regarded as illustrative rather than restrictive, and the specification is not limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.

Claims

1. A method for a user equipment (UE) to communicate in a wireless network, the method comprising: receiving, at the UE, from the wireless network, one or more physical downlink shared channels (PDSCHs) including data of a neural network (NN) model for beam management; verifying, at the UE, the integrity of the NN model received from the wireless network; determining, at the UE, UE capabilities to support the NN model received from the wireless network; as well as In response to verifying the integrity and determining the UE capability to support the NN model, transmitting a NN model transfer completion confirmation to the wireless network.

2. The method according to claim 1, further comprising: generating a compiled NN model by compiling the data of the NN model used by the UE; as well as The compiled NN model is loaded into the NN engine of the UE.

3. The method according to claim 2, wherein transmitting the NN model transfer completion confirmation includes transmitting the NN model transfer completion confirmation from the UE to a server of the wireless network via a base station.

4. The method of claim 3, further comprising, after loading the compiled NN model into the NN engine of the UE, signaling from the UE to the base station that the NN model is ready for use at the UE.

5. The method of claim 2, wherein transmitting the NN model transfer completion confirmation comprises transmitting the NN model transfer completion confirmation to a base station of the wireless network.

6. The method according to claim 2, further comprising: The compiled NN model is generated and loaded into the NN engine of the UE within a time gap signaled to a base station in a UE capability report.

7. The method of claim 1, wherein the NN model is trained by the wireless network for a single cell of the wireless network, and wherein the NN model is associated with a physical cell identifier (PCI).

8. The method of claim 1, wherein the NN model is trained by the wireless network for a single cell of the wireless network, and wherein the NN model is associated with multiple transmit receive points (mTRPs) in the single cell of the wireless network.

9. The method of claim 1, wherein the NN model is trained by the wireless network for multiple cells with different geographic coverage.

10. The method of claim 1, wherein the NN model is trained by the wireless network for multiple cells associated with the same frequency band.

11. The method of claim 1 , further comprising deriving at the UE at least one of a reference signal received power (RSRP) and one or more beam indices using the compiled NN model.

12. The method of claim 1, further comprising receiving, from the wireless network, a configuration for monitoring performance of the NN model.

13. A method for a base station to configure beam management in a wireless network, the method comprising: receiving, at the base station from a server in the wireless network, an indication to load or update a neural network (NN) model for beam management at a user equipment (UE); In response to the indication from the server, sending one or more physical downlink shared channels (PDSCHs) including data for the NN model from the base station to the UE; determining that the UE is ready to use the NN model; and In response to determining that the UE is ready to use the NN model, one or more reference signals (RS) are transmitted to the UE.

14. The method according to claim 13, further comprising: forwarding a NN model transfer completion confirmation from the UE to the server in the wireless network in a transparent manner; as well as A signal is received at the base station from the UE indicating that the UE is ready to use the NN model.

15. The method according to claim 13, further comprising: receiving, at the base station, from the UE a UE capability report indicating a time gap value; receiving, at the base station, a NN model transfer completion confirmation from the UE; In response to receiving the NN model transfer completion confirmation, starting a timer corresponding to the time gap value; as well as When the timer expires, it is determined that the UE is ready to use the NN model.

16. The method of claim 13, wherein the NN model is trained by the server in the wireless network for a single cell of the wireless network, and wherein the NN model is associated with a physical cell identifier (PCI).

17. The method of claim 13, wherein the NN model is trained by the server in the wireless network for a single cell of the wireless network, and wherein the NN model is associated with multiple transmit receive points (mTRPs) in the single cell of the wireless network.

18. The method of claim 13, wherein the NN model is trained by the server in the wireless network for multiple cells with different geographic coverage.

19. The method of claim 13, wherein the NN model is trained by the server in the wireless network for multiple cells associated with the same frequency band.

20. The method of claim 13, further comprising transmitting, from the base station to the UE, a configuration for monitoring the performance of the NN model.

21. An apparatus comprising means for performing the method according to any one of claims 1 to 20.

22. A computer-readable medium comprising instructions, which, when executed by one or more processors of an electronic device, cause the electronic device to perform the method according to any one of claims 1 to 20.

23. An apparatus comprising logic components, modules or circuits for performing the method according to any one of claims 1 to 20.

Images