Performance monitoring for TSF domain CSI compression

Abstract

A technique performed by a user equipment, UE, supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression is disclosed. A method implementation of the technique comprises generating a performance monitoring result in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring, and transmitting the generated performance monitoring result to a network node.

Description

Telefonaktiebolaget LM Ericsson (publ) - 1 - 30A-169 639PERFORMANCE MONITORING FORTSF DOMAIN CSI COMPRESSIONTECHNICAL FIELD[OOO1] The present disclosure generally relates to wireless communication systems. In particular, a technique for performance monitoring for Temporal-Spatial-Frequency (TSF) domain Channel State Information (CSI) compression is presented. The technique may be embodied in methods, apparatuses, computer programs and systems.BACKGROUND

[0002] Artificial Intelligence (Al) and Machine Learning (ML) have been investigated, both in academia and industry, as promising tools to optimize the air interface design in wireless communication networks. One example of AI / ML implementation for the physical layer (Al PHY) is using an AI / ML autoencoder to improve the channel compression accuracy and / or to reduce the channel state information (CSI) feedback overhead. This topic has been discussed throughout 3rd Generation Partnership Project (3GPP) Rel. 18 and continues to be discussed in 3GPP Rel. 19 systems.General aspects for AI / ML-based CSI compression

[0003] In the legacy mechanism (i.e., non-AI / ML-based CSI compression), the user equipment (UE) can be configured to report a suggested precoder to the network (NW), a so-called precoder matrix indicator (PMI). The PMI is based on a measured CSI-Reference Signal (CSI-RS) and sent to the NW as a CSI report, based on a certain mechanism, the so-called codebook. The codebook generally defines how the UE arranges the reported bits based on (the number of) beams and taps selected by the UE to be reported to the NW and how the UE quantizes the precoding matrix.

[0004] In AI / ML-based CSI compression, an AI / ML-based autoencoder (AE) replaces at least part of the legacy mechanism. AEs can have different architectures. For example, AEs can be based on dense neural networks (NNs), multi-dimensional convolution NNs, variational, recurrent NNs, transformer networks, or any combination thereof. However, all AE architectures possess an encoder-bottleneck-decoder structure illustrated in Figure 1.

[0005] The codeword's size (denoted by Y in Figure 1) of an AE is smaller than the input data's size (X in Figure 1). The AE encoder thus reduces the dimensionality of the input features X down to Y's. The decoder part of the AE tries to invert theTelefonaktiebolaget LM Ericsson (publ) - 2 - 30A-169 639encoder and reconstruct X with minimal error, according to some predefined loss function, also known as the target function in the general optimization literature.

[0006] Figure 2 illustrates how an AE might be used for AI / ML-enhanced CSI reporting in wireless communication systems, such as new radio (NR). The UE measures the channel in the downlink using CSI-RS. The UE estimates the channel for each subcarrier (SC) from each base station (BS) transmitter (TX) antenna to each UE receiver (RX) antenna. The estimate can be viewed as a three-dimensional (3D) channel matrix. The 3D channel matrix represents the multiple-input multiple-output (MIMO) channel estimated over several SCs and is input to the encoder. However, there are different architectures where a processed version of the 3D MIMO channel, or a processed subset of the information, e.g., singular vectors of the MIMO channel, is input to the encoder.

[0007] The AE encoder is implemented in the UE, and the AE decoder is implemented in the NW, denoted "BS" for base station in Figure 2. The output of the AE encoder is signaled from the UE to the NW over the uplink. The codeword can be viewed as a learned latent representation of the channel. Properties of the data (e.g., CSI-RS channel estimates), the channel size, uplink feedback rate, and hardware limitations of the encoder and decoder need to be considered when optimizing the AE's architecture.

[0008] The weights and biases of an AE (with a fixed architecture) are trained to minimize the reconstruction error (the error between the input X and output X) on some training datasets. For example, the weights and biases can be trained to minimize the mean squared error, MSE (mean(x - X) ). To achieve good performance during live operation, the training data set should represent the actual data the AE will encounter during live operation.

[0009] In the two-sided CSI compression, the output of the UE-side encoder needs to be communicated over the air interface to the BS (e.g., gNodeB (gNB) in NR) decoder with the assigned CSI reporting payload and, therefore, needs to be quantized to a finite number of bits (e.g., 1-4 bits per encoder output's neuron) to obtain an efficient transmission as shown in Figure 3. Accordingly, a quantization layer is connected to the output of the encoder or directly included in the encoder. In an example, the quantization layer may implement scalar quantization which quantizes the output of each neuron of the encoder output layer (the bottleneck layer of AE) to generate bits to fit the CSI reporting payload in the uplink control information (UCI). Other quantization methods, e.g., vector quantization, may also be used.Telefonaktiebolaget LM Ericsson (publ) - 3 - 30A-169 639Pre-processing for input data to the AE

[0010] A proper pre-processing on the input to the encoder can greatly reduce the size and complexity of designing and / or training an AI / ML model, and, in the meantime, improve the scalability and transferability of the model. In addition, preprocessing may reduce the need for multiple models depending on bandwidth variation and variation in the number of antenna ports at the BS. By using pre-processing, instead of directly compressing the channel (i.e., with dimensions of RX x TX x SC as in Figure 2), the channels are first processed into another representation.

[0011] One example of the pre-processing may include transforming the channel into eigenvectors as shown in Figure 4. Here, the UE may conduct the following steps: 1. Compute the covariance matrix of the channel and extract the relevant eigenvectors.2. The covariance matrix is summed over 4 "f-units", to get 13 "sub-chunks" (subbands) in frequency.3. For each of the 13 averaged covariance matrices, compute an eigen-decomposition and extract the 4 eigenvectors corresponding to the 4 largest eigenvalues.4. Normalize the phase and magnitude of the eigenvectors.5. The number of eigenvectors to feedback should be the same for all sub-chunks and depends on the rank hypothesis testing with a value between 1 and 4.

[0012] In another example, pre-processing may include transforming the channel into the beam-delay domain. The feature extraction for beam-delay reduced eigenvector-based feedback is exemplarily illustrated in Figure 5. This is also called a "W2 compressor" as the feature extraction is similar to the standardized 3GPP procedures and the jargon for the matrix left to compress for the AI / ML is "W2". The steps are as follows:1. The UE selects an orthogonal basis, one of a set of oversa mpled / rotated, spatial domain, discrete Fourier transform (DFT) bases. On that basis, the UE selects L vectors. These vectors represent the spatial domain (SD) basis DFT. This basis is wideband applicable, i.e., valid for all subbands, and applicable for both polarizations. This selection may happen jointly, e.g., the UE does a spatial domain DFT on the 32 x 4 (TX x RX) matrix, per resource block (RB), and selects the L strongest beams out of 16 (for one polarization). The beam-space channel is computed by multiplying the channel with the selected SD basis. The covariance of the beam-space channel is summed over, e.g., 4 RBs to produce a covariance matrix for each subband.2. For each covariance matrix (per subband), the UE extracts a number of eigenvectors and may select the rank, i.e., number of layers.3. The UE does a frequency domain DFT per layer, transforming to a delay domain, whereafter it selects the M strongest taps. The resulting tensor of dimensions 2L x number of layers x M is called the linear combination coefficients, in 3GPP jargon "W2".Telefonaktiebolaget LM Ericsson (publ) - 4 - 30A-169 639The W2 matrix can be used to reconstruct the, by the UE suggested, precoding matrices.4. The data is used as input in the AI / ML model. This could be the raw linear combination coefficients, or it could be enhanced with information about the selected beams and taps, noise levels, etc.Performance monitoring for Al-based spatial-frequency (SF) domain CSI compression

[0013] To make sure that the performance of the AI / ML is acceptable, monitoring the performance of the AI / ML model is essential. Note that, here, the term acceptable may not only be in the form of absolute value but also in the form of relative value, e.g., by comparing the performance with the (expected) performance of the legacy mechanism. Performance monitoring can be done by using an intermediate key performance indicator (KPI), e.g., normalized mean square error (NMSE), squared generalized cosine similarity (SGCS), etc.; eventual KPI, e.g., user perceived throughput (UPT), the expected block error rate (BLER), etc.; or by monitoring the data drift either in the AI / ML input or the AI / ML output (or both). Performance monitoring can be done either on the UE side or the NW side (or both).UE-side monitoring

[0014] In UE-side monitoring, a UE can be configured to measure one or more channel samples and transmit the performance metric of the (UE part) model for the respective channels. In one example, the UE can achieve this by measuring the channel, conducting the required preprocessing, conducting the model inference, and using the output of the UE part model to a nominal decoder. The output of the nominal decoder is then compared to the ground-truth / target-CSI (i.e., the measured channels or the pre-processed measured channels). The nominal decoder may be a reference decoder agreed in the standard, a decoder provided by the NW, a decoder produced by the UE during the model training, etc. In another alternative, the UE may also train a performance metric estimator model which may use the measured channels as the input and directly output the (estimated) performance metric of the UE part model.NW-side monitoring

[0015] To achieve reliable model performance assessment results, the UE can be configured to measure one or more channel samples and then report the target-CSI and the model output(s) of one or multiple UE-part model(s) associated with these one or more channel samples. In the case in which multiple samples of channels are used, the UE may first accumulate the target-CSI and the model output(s) for multipleTelefonaktiebolaget LM Ericsson (publ) - 5 - 30A-169 639samples within a time window and then report the accumulated data together, or the UE may report the target-CSI and model output(s) per sample.

[0016] This target-CSI, in return, may be used by the NW to do performance monitoring, model retraining, fine-tuning, etc. The NW, may, for example, first generate one or more reconstructed CSI samples by feeding the one or more UE-part model output samples as inputs to the NW-part model, and then calculate one or more intermediate KPI values by comparing the one or more generated reconstructed CSI with the target CSI samples. Some of the possible intermediate KPIs, for example, are the generalized cosine similarity (GCS), squared generalized cosine similarity (SGCS), etc.

[0017] To achieve higher accuracy in performance monitoring, data collection, etc., the target-CSI needs to be as accurate as possible. However, this may bring a significant amount of overhead. Therefore, the target-CSI may need to be quantized to reduce the overhead. The quantization may be done, e.g., with scalar quantization, with e-Type Il-like quantization, etc.Al-based TSF domain CSI compression with predicted CSI

[0018] The 3GPP Rel. 18 study on Al CSI compression focused on the compression of a channel measurement in the spatial and frequency domain. In 3GPP Rel. 19, new use cases have been considered for CSI compression to also include the temporal domain compression aspects. One mechanism to consider the temporal domain of CSI compression is by compressing multiple CSI reports at once. The rationale is that, given that the reports are close in time, they could have correlations to be learned from the Al model to reach greater levels of compression. The caveat in this case is that CSI measurements of previous slots are not readily useful for future transmissions at the base station, thus, the CSI reports must be for the future. Thus, CSI prediction at the UE is introduced.

[0019] Figure 6 illustrates an example implementation. K previous channel measurements are used to predict CSI of / V4future slots. The / V4slots are then fed to an encoder which transforms all the measurements into a single report, with a size smaller than its input. Such self-contained report is then forwarded to the NW which uses the decoder to restore the report for the / V4slots. As with the previous illustrated examples, other processes can be introduced, such as pre-processing, and different AI / ML methods can be used.

[0020] Note that prediction and compression can be seen as logical processes, and they can be implemented separately or combined in a single AI / ML model or even be implemented by non-AI, such as autoregressive models for prediction. Figure 7 shows an example in which prediction and compression are implemented jointly.Telefonaktiebolaget LM Ericsson (publ) - 6 - 30A-169 639

[0021] There currently exist certain challenges. One challenge in the performance monitoring for TSF domain CSI compression lies in how to identify the root cause of the performance degradation, i.e., whether the performance degradation comes from the CSI compression part or comes from the insufficient quality of the prediction block output, for example.SUMMARY

[0022] Certain aspects of the disclosure and their embodiments may provide solutions to this or other challenges. To address this challenge, performance monitoring could be done based on several possible target-CSI resources, e.g., based on the actual predicted CSI (e.g., predicted precoder or channel) obtained from the output of the (logical) CSI prediction block (also called "predicted CSI" herein), or based on the "ideally predicted CSI" obtained from CSI measurement from the CSI-RS in the prediction window (also called "measured CSI" herein). Such two different target-CSI types entail different requirements, configurations, indications, and reporting formats which need to be defined.

[0023] In the present disclosure, mechanisms for performance monitoring for (e.g., Al-based) TSF domain CSI compression with predicted CSI are described, specifically, to support multiple types of target-CSI formats. In particular, the present disclosure covers UE capability aspects regarding performance monitoring, performance monitoring configurations, the format and mapping of the performance monitoring indication from the NW to the UE, and the performance monitoring format from the UE to the NW of at least one of performance monitoring reports or target-CSI reports. Features from certain aspects described in the following may be combined with other aspects disclosed herein. The same applies to different embodiments disclosed herein.

[0024] According to a first aspect, a method performed by a user equipment (UE) supporting performance monitoring for Temporal-Spatial-Frequency (TSF) domain Channel State Information (CSI) compression is provided. The method comprises generating a performance monitoring result in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring. The method further comprises transmitting the generated performance monitoring result to a network node.

[0025] The performance monitoring result may comprise one of a target CSI report and a performance monitoring report generated by the UE. The method may further comprise transmitting, to the network node, UE capability information indicating at least one capability of the UE in relation to the performance monitoring. The UE capability information may comprise one or more target CSI types supportedTelefonaktiebolaget LM Ericsson (publ) - 7 - 30A-169 639by the UE for the performance monitoring. The one or more target CSI types may include at least one of a measured CSI obtained by the UE from CSI measurement of a CSI-Reference Signal (CSI-RS) in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

[0026] The UE capability information may comprise an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI. The one or more CSI prediction mechanisms may include at least one of an Artificial Intelligence (Al) based CSI prediction mechanism and a non-AI based CSI prediction mechanism. The UE capability information may comprise an indication of a processing time required to complete a performance monitoring operation. The indication of the processing time may be made per target CSI type. The indication of the processing time may be made per performance monitoring type. The UE capability information may comprise an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly. When the UE is configured to perform CSI prediction and CSI compression separately, the UE capability information may comprise an indication that the UE supports CSI prediction performance monitoring. The indication that the UE supports CSI prediction performance monitoring may be a prerequisite for a capability of the UE for performing TSF compression.

[0027] The at least one parameter defined by the performance monitoring configuration information may comprise a target CSI type to be used for the performance monitoring. The target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE. When the target CSI type is a predicted CSI, the target CSI type may be one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms. The underlying CSI prediction mechanisms include at least an Al based CSI prediction mechanism and a non-AI based CSI prediction mechanism.

[0028] In the performance monitoring configuration information, the target CSI type may be defined per performance monitoring type. A performance monitoring periodicity may be defined for each target CSI type in the performance monitoring configuration information. In one variant, the method may further comprise receiving the performance monitoring configuration information from the network node. In another variant, the method may further comprise deriving the performance monitoring configuration information from a configuration for a CSI-RS in a prediction window for performance monitoring.

[0029] The method may further comprise receiving, from the network node, an indication to conduct performance monitoring. The indication may include the performance monitoring configuration information as a selection among availableTelefonaktiebolaget LM Ericsson (publ) - 8 - 30A-169 639configuration options configured for the UE. The indication may be made via a bitfield contained in control information received by the UE from the network node. The control information comprises one of a Downlink Control Information (DCI) and a Media Access Control (MAC) Control Element (CE). A size of the bitfield may be dependent on a number of target CSI types configured for the UE. Bits of the bitfield may be interpreted in accordance with at least one of a meaning defined in a standard, a network configuration, and a bitmap representation in which each bit is assigned a particular meaning.

[0030] When the performance monitoring result comprises a target CSI report, the performance monitoring result may further comprise an indication of a target CSI type based on which the target CSI report is generated. The performance monitoring result may comprise a plurality of target CSI reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated. When the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result may further comprise an indication of a target CSI type based on which the performance monitoring report is generated. The performance monitoring result may comprise a plurality of performance monitoring reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated. Target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

[0031] According to a second aspect, a method performed by a network node supporting performance monitoring for Temporal-Spatial-Frequency (TSF) domain Channel State Information (CSI) compression is provided. The method comprises receiving a performance monitoring result from a user equipment (UE), wherein the performance monitoring result is generated in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring. The method according to the second aspect may define a method from the perspective of a network node described above in relation to the method according to the first aspect.

[0032] The performance monitoring result may comprise one of a target CSI report and a performance monitoring report generated by the UE. The method may further comprise receiving, from the UE, UE capability information indicating at least one capability of the UE in relation to the performance monitoring. The UE capability information may comprise one or more target CSI types supported by the UE for theTelefonaktiebolaget LM Ericsson (publ) - 9 - 30A-169 639performance monitoring. The one or more target CSI types may include at least one of a measured CSI obtained by the UE from CSI measurement of a CSI-Reference Signal (CSI-RS) in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

[0033] The UE capability information may comprise an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI. The one or more CSI prediction mechanisms may include at least one of an Artificial Intelligence (Al) based CSI prediction mechanism and a non-AI based CSI prediction mechanism. The UE capability information may comprise an indication of a processing time required to complete a performance monitoring operation. The indication of the processing time may be made per target CSI type. The indication of the processing time may be made per performance monitoring type. The UE capability information may comprise an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly. When the UE is configured to perform CSI prediction and CSI compression separately, the UE capability information may comprise an indication that the UE supports CSI prediction performance monitoring. The indication that the UE supports CSI prediction performance monitoring may be a prerequisite for a capability of the UE for performing TSF compression.

[0034] The at least one parameter defined by the performance monitoring configuration information may comprise a target CSI type to be used for the performance monitoring. The target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE. When the target CSI type is a predicted CSI, the target CSI type may be one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms. The underlying CSI prediction mechanisms may include at least an Al based CSI prediction mechanism and a non-AI based CSI prediction mechanism.

[0035] In the performance monitoring configuration information, the target CSI type may be defined per performance monitoring type. A performance monitoring periodicity may be defined for each target CSI type in the performance monitoring configuration information. The method may further comprise transmitting the performance monitoring configuration information to the UE.

[0036] The method may further comprise transmitting, to the UE, an indication to conduct performance monitoring. The indication may include the performance monitoring configuration information as a selection among available configuration options configured for the UE. The indication may be made via a bitfield contained in control information transmitted from the network node to the UE. The control information may comprise one of a Downlink Control Information (DCI) and a MediaTelefonaktiebolaget LM Ericsson (publ) - 10 - 30A-169 639Access Control (MAC) Control Element (CE). A size of the bitfield may be dependent on a number of target CSI types configured for the UE. Bits of the bitfield may be interpreted in accordance with at least one of a meaning defined in a standard, a network configuration, and a bitmap representation in which each bit is assigned a particular meaning.

[0037] When the performance monitoring result comprises a target CSI report, the performance monitoring result may further comprise an indication of a target CSI type based on which the target CSI report is generated. The performance monitoring result may comprise a plurality of target CSI reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated. When the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result may further comprise an indication of a target CSI type based on which the performance monitoring report is generated. The performance monitoring result may comprise a plurality of performance monitoring reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated. The target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

[0038] According to a third aspect, a computer program product is provided. The computer program product comprises program code portions for performing the method of the first aspect when the computer program product is executed on a user equipment (UE). The computer program product may be stored on a computer readable recording medium, such as a semiconductor memory, DVD, CD-ROM, and so on.

[0039] According to a fourth aspect, a computer program product is provided. The computer program product comprises program code portions for performing the method of the second aspect when the computer program product is executed on a network node. The computer program product may be stored on a computer readable recording medium, such as a semiconductor memory, DVD, CD-ROM, and so on.

[0040] According to a fifth aspect, a user equipment (UE) supporting performance monitoring for Temporal-Spatial-Frequency (TSF) domain Channel State Information (CSI) compression is provided. The UE comprises processing circuitry configured to perform any of the steps of method of the first aspect.

[0041] According to a sixth aspect, a network node supporting performance monitoring for Temporal-Spatial-Frequency (TSF) domain Channel State InformationTelefonaktiebolaget LM Ericsson (publ) - 11 - 30A-169 639(CSI) compression is provided. The network node comprises processing circuitry configured to perform any of the steps of method of the second aspect.

[0042] According to an seventh aspect, there is provided a system comprising a user equipment (UE) of the fifth aspect and a network node of the sixth aspect.

[0043] Certain embodiments and / or aspects may provide one or more of the following technical advantages. Using the techniques presented herein, the UE and the NW may have a same understanding of which performance monitoring approaches (e.g., measured-CSI-based or predicted-CSI-based) the UE is capable of, and which performance monitoring approaches shall be done, or are currently reported by the UE. The use of several possible target-CSI resources also enables the NW and / or the UE to conduct root cause analysis, should a degradation in the performance occur, as well as to trigger life cycle management (LCM) procedures accordingly.BRIEF DESCRIPTION OF THE DRAWINGSFig. 1 illustrates an example of a fully connected autoencoder;Fig. 2 illustrates an exemplary scheme of using an autoencoder for CSI compression;Fig. 3 illustrates an exemplary quantization operation at the output of an encoder to fit the CSI payload over the air interface;Fig. 4 illustrates an exemplary scheme of pre-processing a channel into eigenvectors before using it as inputs to an encoder;Fig. 5 illustrates an exemplary scheme of pre-processing a channel into a beam delay domain before using it as inputs to an encoder;Fig. 6 illustrates an example of CSI compression with predicted CSI, wherein a separate block with channel / CSI block prediction is used;Fig. 7 illustrates an example of CSI compression with predicted CSI, wherein a combined block with channel / CSI block prediction and compression is used;Fig. 8 illustrates an exemplary system in accordance with the present disclosure;Fig. 9 illustrates an exemplary UE in accordance with the present disclosure; Fig. 10 illustrates an exemplary network node in accordance with the present disclosure;Fig. 11 illustrates a method which may be performed by a UE according to the present disclosure;Fig. 12 illustrates a method which may be performed by a network node according to the present disclosure; andTelefonaktiebolaget LM Ericsson (publ) - 12 - 30A-169 639Fig. 13 illustrates a signaling diagram of an exemplary interaction between the UE and an node according to the present disclosure.DETAILED DESCRIPTION

[0044] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0045] Figure 8 shows an example of a communication system 800. In the example, the communication system 800 includes a telecommunication network 802 that includes an access network 804, such as a radio access network (RAN), and a core network 806, which includes one or more core network nodes 808. The access network 804 includes one or more access network nodes, such as network nodes 810a and 810b (one or more of which may be generally referred to as network nodes 810), or any other similar 3rd Generation Partnership Project (3GPP) access nodes or non-3GPP access points. Moreover, as will be appreciated by those of skill in the art, a network node is not necessarily limited to an implementation in which a radio portion and a baseband portion are supplied and integrated by a single vendor. Thus, it will be understood that network nodes include disaggregated implementations or portions thereof. For example, in some embodiments, the telecommunication network 802 includes one or more Open-RAN (ORAN) network nodes. An ORAN network node is a node in the telecommunication network 802 that supports an ORAN specification (e.g., a specification published by the O-RAN Alliance, or any similar organization) and may operate alone or together with other nodes to implement one or more functionalities of any node in the telecommunication network 802, including one or more network nodes 810 and / or core network nodes 808. Examples of an ORAN network node include an open radio unit (O-RU), an open distributed unit (O-DU), and an open central unit (O-CU).

[0046] The network nodes 810 facilitate direct or indirect connection of user equipment (UE), such as by connecting UEs 812a, 812b and 812c (one or more of which may be generally referred to as UEs 812) to the core network 806 over one or more wireless connections. Example wireless communications over a wireless connection include transmitting and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for conveying information without the use of wires, cables, or other material conductors. Moreover, in different embodiments, the communication system 800 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that may facilitate or participate in the communication of dataTelefonaktiebolaget LM Ericsson (publ) - 13 - 30A-169 639and / or signals whether via wired or wireless connections. The communication system 800 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar type of system.

[0047] The UEs 812 may be any of a wide variety of communication devices, including wireless devices arranged, configured, and / or operable to communicate wirelessly with the network nodes 810 and other communication devices. Similarly, the network nodes 810 are arranged, capable, configured, and / or operable to communicate directly or indirectly with the UEs 812 and / or with other network nodes or equipment in the telecommunication network 802 to enable and / or provide network access, such as wireless network access, and / or to perform other functions, such as administration in the telecommunication network 802.

[0048] In the depicted example, the core network 806 includes one more core network nodes (e.g., core network node 808) that are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, network nodes, and / or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node 808. Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (AUSF), Subscription Identifier De-concealing function (SIDF), Unified Data Management (UDM), Security Edge Protection Proxy (SEPP), Network Exposure Function (NEF), and / or a User Plane Function (UPF).

[0049] As a whole, the communication system 800 of Figure 8 enables connectivity between the UEs, network nodes, and hosts. In that sense, the communication system may be configured to operate according to predefined rules or procedures, such as specific standards that include, but are not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and / or other suitable 2G, 3G, 4G, 5G standards, or any applicable future generation standard (e.g., 6G); wireless local area network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards (WiFi); and / or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and / or any low-power wide-area network (LPWAN) standards such as LoRa and Sigfox.

[0050] Figure 9 shows a UE 900 in accordance with some embodiments. The UE 900 presents additional details of some embodiments of the UE 812 of Figure 8. As used herein, a UE refers to a device capable, configured, arranged and / or operable toTelefonaktiebolaget LM Ericsson (publ) - 14 - 30A-169 639communicate wirelessly with network nodes and / or other UEs. Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage / playback device, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), la ptop- mounted equipment (LME), an Augmented Reality (AR) or Virtual Reality (VR) device, wireless customer-premise equipment (CPE), vehicle, vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-IoT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0051] The UE 900 includes processing circuitry 902 that is operatively coupled to a memory 910, a communication interface 912, and / or any other component, or any combination thereof. Certain UEs may utilize all or a subset of the components shown in Figure 9. The level of integration between the components may vary from one UE to another UE. Further, certain UEs may contain multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0052] The processing circuitry 902 is configured to process instructions and data and may be configured to implement any sequential state machine operative to execute instructions stored as machine-readable computer programs in the memory 910. The processing circuitry 902 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), etc.); programmable logic together with appropriate firmware; one or more stored computer programs, general-purpose processors, such as a microprocessor or digital signal processor (DSP), together with appropriate software; or any combination of the above. For example, the processing circuitry 902 may include multiple central processing units (CPUs).

[0053] The memory 910 may be or be configured to include memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and so forth. In one example, the memory 910 includes one or more application programs, such as an operating system, web browser application, a widget, gadget engine, or other application, and corresponding data. The memory 910 may store, for use by the UE 900, any of a variety of various operating systems or combinations of operating systems.Telefonaktiebolaget LM Ericsson (publ) - 15 - 30A-169 639

[0054] The memory 910 may be configured to include a number of physical drive units, such as redundant array of independent disks (RAID), flash memory, USB flash drive, external hard disk drive, thumb drive, pen drive, key drive, high-density digital versatile disc (HD-DVD) optical disc drive, internal hard disk drive, Blu-Ray optical disc drive, holographic digital data storage (HDDS) optical disc drive, external mini-dual inline memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro-DIMM SDRAM, smartcard memory such as tamper resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs), such as a USIM and / or ISIM, other memory, or any combination thereof. The UICC may for example be an embedded UICC (eUICC), integrated UICC (iUICC) or a removable UICC commonly known as 'SIM card.' The memory 910 may allow the UE 900 to access instructions, application programs and the like, stored on transitory or non-transitory memory media, to off-load data, or to upload data. An article of manufacture, such as one utilizing a communication system may be tangibly embodied as or in the memory 910, which may be or comprise a device-readable storage medium.

[0055] The processing circuitry 902 may be configured to communicate with an access network or other network using the communication interface 912. The communication interface 912 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna 922. The communication interface 912 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in an access network). Each transceiver may include a transmitter 918 and / or a receiver 920 appropriate to provide network communications (e.g., optical, electrical, frequency allocations, and so forth). Moreover, the transmitter 918 and receiver 920 may be coupled to one or more antennas (e.g., antenna 922) and may share circuit components, software or firmware, or alternatively be implemented separately.

[0056] Figure 10 shows a network node 1000 in accordance with some embodiments. As used herein, network node refers to equipment capable, configured, arranged and / or operable to communicate directly or indirectly with a UE and / or with other network nodes or equipment, in a telecommunication network. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)), O-RAN nodes or components of an O-RAN node (e.g., O-RU, O-DU, O-CU).

[0057] Base stations may be categorized based on the amount of coverage they provide (or, stated differently, their transmit power level) and so, depending on theTelefonaktiebolaget LM Ericsson (publ) - 16 - 30A-169 639provided amount of coverage, may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations. A base station may be a relay node or a relay donor node controlling a relay. A network node may also include one or more (or all) parts of a distributed radio base station such as centralized digital units, distributed units (e.g., in an O-RAN access node) and / or remote radio units (RRUs), sometimes referred to as Remote Radio Heads (RRHs). Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio. Parts of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).

[0058] Other examples of network nodes include multiple transmission point (multi-TRP) 5G access nodes, multi-standard radio (MSR) equipment such as MSR BSs, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base transceiver stations (BTSs), transmission points, transmission nodes, multi-cell / multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, Self-Organizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and / or Minimization of Drive Tests (MDTs).

[0059] The network node 1000 includes a processing circuitry 1002, a memory 1004, and a communication interface 1006. The network node 1000 may be composed of multiple physically separate components (e.g., a NodeB component and a RNC component, or a BTS component and a BSC component, etc.), which may each have their own respective components. In certain scenarios in which the network node 1000 comprises multiple separate components (e.g., BTS and BSC components), one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple NodeBs. In such a scenario, each unique NodeB and RNC pair, may in some instances be considered a single separate network node. In some embodiments, the network node 1000 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memory 1004 for different RATs) and some components may be reused (e.g., a same antenna 1010 may be shared by different RATs). The network node 1000 may also include multiple sets of the various illustrated components for different wireless technologies integrated into network node 1000, for example GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, Radio Frequency Identification (RFID) or Bluetooth wireless technologies. These wireless technologies may be integrated into the same or different chip or set of chips and other components within network node 1000.

[0060] The processing circuitry 1002 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digitalTelefonaktiebolaget LM Ericsson (publ) - 17 - 30A-169 639signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and / or encoded logic operable to provide, either alone or in conjunction with other network node 1000 components, such as the memory 1004, to provide network node 1000 functionality.

[0061] In some embodiments, the processing circuitry 1002 includes a system on a chip (SOC). In some embodiments, the processing circuitry 1002 includes one or more of radio frequency (RF) transceiver circuitry 1012 and baseband processing circuitry 1014. In some embodiments, the radio frequency (RF) transceiver circuitry 1012 and the baseband processing circuitry 1014 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry 1012 and baseband processing circuitry 1014 may be on the same chip or set of chips, boards, or units.

[0062] The memory 1004 may comprise any form of volatile or non-volatile computer-readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and / or any other volatile or non-volatile, non-transitory device-readable and / or computer-executable memory devices that store information, data, and / or instructions that may be used by the processing circuitry 1002. The memory 1004 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and / or other instructions capable of being executed by the processing circuitry 1002 and utilized by the network node 1000. The memory 1004 may be used to store any calculations made by the processing circuitry 1002 and / or any data received via the communication interface 1006. In some embodiments, the processing circuitry 1002 and memory 1004 is integrated.

[0063] The communication interface 1006 is used in wired or wireless communication of signaling and / or data between a network node, access network, and / or UE. As illustrated, the communication interface 1006 comprises port(s) / terminal(s) 1016 to send and receive data, for example to and from a network over a wired connection. The communication interface 1006 also includes radio frontend circuitry 1018 that may be coupled to, or in certain embodiments a part of, the antenna 1010. Radio front-end circuitry 1018 comprises filters 1020 and amplifiers 1022. The radio front-end circuitry 1018 may be connected to an antenna 1010 and processing circuitry 1002. The radio front-end circuitry may be configured to condition signals communicated between antenna 1010 and processing circuitry 1002. The radioTelefonaktiebolaget LM Ericsson (publ) - 18 - 30A-169 639front-end circuitry 1018 may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection. The radio front-end circuitry 1018 may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters 1020 and / or amplifiers 1022. The radio signal may then be transmitted via the antenna 1010. Similarly, when receiving data, the antenna 1010 may collect radio signals which are then converted into digital data by the radio front-end circuitry 1018. The digital data may be passed to the processing circuitry 1002. In other embodiments, the communication interface may comprise different components and / or different combinations of components.

[0064] In certain alternative embodiments, the network node 1000 does not include separate radio front-end circuitry 1018, instead, the processing circuitry 1002 includes radio front-end circuitry and is connected to the antenna 1010. Similarly, in some embodiments, all or some of the RF transceiver circuitry 1012 is part of the communication interface 1006. In still other embodiments, the communication interface 1006 includes one or more ports or terminals 1016, the radio front-end circuitry 1018, and the RF transceiver circuitry 1012, as part of a radio unit (not shown), and the communication interface 1006 communicates with the baseband processing circuitry 1014, which is part of a digital unit (not shown).

[0065] The antenna 1010 may include one or more antennas, or antenna arrays, configured to send and / or receive wireless signals. The antenna 1010 may be coupled to the radio front-end circuitry 1018 and may be any type of antenna capable of transmitting and receiving data and / or signals wirelessly. In certain embodiments, the antenna 1010 is separate from the network node 1000 and connectable to the network node 1000 through an interface or port.

[0066] The antenna 1010, communication interface 1006, and / or the processing circuitry 1002 may be configured to perform any receiving operations and / or certain obtaining operations described herein as being performed by the network node. Any information, data and / or signals may be received from a UE, another network node and / or any other network equipment. Similarly, the antenna 1010, the communication interface 1006, and / or the processing circuitry 1002 may be configured to perform any transmitting operations described herein as being performed by the network node. Any information, data and / or signals may be transmitted to a UE, another network node and / or any other network equipment.

[0067] Embodiments of the network node 1000 may include additional components beyond those shown in Figure 10 for providing certain aspects of the network node's functionality, including any of the functionality described herein and / or any functionality necessary to support the subject matter described herein. For example, the network node 1000 may include user interface equipment to allow inputTelefonaktiebolaget LM Ericsson (publ) - 19 - 30A-169 639of information into the network node 1000 and to allow output of information from the network node 1000. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node 1000. In some embodiments providing a core network node, such as core network node 108 of FIG. 8, some components, such as the radio front-end circuitry 1018 and the RF transceiver circuitry 1012 may be omitted.

[0068] Although the computing devices described herein (e.g., UEs, network nodes) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.

[0069] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer-readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry aloneTelefonaktiebolaget LM Ericsson (publ) - 20 - 30A-169 639or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

[0070] Figure 11 illustrates a method which may be performed by a UE according to the present disclosure. The UE supports performance monitoring for Temporal-Spatial-Frequency (TSF) domain Channel State Information (CSI) compression. In step SI 102, the UE generates a performance monitoring result in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring. In step 1104, the UE transmits the generated performance monitoring result to a network node. The UE may correspond to the UE 900 and the processing circuitry 902 of the UE may be configured such that the UE is operable to carry out the method steps described herein with respect to the UE.

[0071] The UE may be capable of implementing TSF domain CSI compression in a way described above in relation to 3GPP Rel. 18 and 3GPP Rel. 19 systems, respectively. The UE may as such be capable of implementing Artificial Intelligence (Al) based TSF domain CSI compression and its corresponding performance monitoring features. In this regard, the UE (and the NW in two-sided model cases) may be assumed to have the capability of running AI / ML models supporting corresponding AI / ML enabled features and their respective performance monitoring procedures described above. The UE and the network node may be configured to perform at least one of "NW-side monitoring" (i.e., network-side performance monitoring) and "UE-side monitoring" (i.e., UE-side performance monitoring). In certain variants, the determination whether the performance monitoring is UE-side or NW-side may be configurable, e.g., by a configuration message sent from the network node to the UE or from the UE to the network node. The term "NW" ("network") as used herein may mean any network node within the meaning described herein, e.g., including a generic network node, gNB, base station, unit within the base station, relay node, core network node, device supporting device-to-device (D2D) communication, or the like.

[0072] As described above, in UE-side performance monitoring, a UE may be configured to measure one or more channel samples and transmit a performance metric (e.g., a desired target key performance indicator (KPI)) of a UE part model for the respective channels. To this end, the UE may measure the channel, conduct required pre-processing, conduct the model inference, and use the output of the model to a nominal decoder, for example. The output of the nominal decoder may then be compared to the target CSI (i.e., the measured channels or the pre-processed measured channels), which may be taken as the ground-truth. The nominal decoder may be a reference decoder defined in a standard, a decoder provided by the NW, a decoder produced by the UE during the model training, or the like. As an alternative,Telefonaktiebolaget LM Ericsson (publ) - 21 - 30A-169 639the UE may also train a performance metric estimator model which may use the measured channels as input and may directly output the (estimated) performance metric of the UE part model.

[0073] As also described above, in NW-side performance monitoring, the UE may be configured to measure one or more channel samples and then report the target CSI and the model outputs of one or more UE part models associated with these one or more channel samples. If multiple samples of channels are used, the UE may first accumulate the target CSI and the model outputs for one or more samples within a time window and then report the accumulated data together, or the UE may report the target CSI and model outputs per sample. This target CSI, in turn, may be used by the network node to conduct performance monitoring, model retraining, fine-tuning, or the like. The network node may, for example, first generate one or more reconstructed CSI samples by feeding the one or more UE part model output samples as inputs to the NW-part model, and then calculate one or more performance metric (e.g., desired target KPI) values by comparing the one or more generated reconstructed CSI with the target CSI samples. The target CSI may be quantized to reduce overhead, wherein such quantization may be performed using scalar quantization, e-Type Il-like quantization, or the like.

[0074] In the method performed by the UE, a performance monitoring result is generated in accordance with the performance monitoring configuration information (step S1102). Such performance monitoring configuration information may be a dedicated configuration applied to the UE, wherein the configuration may define one or more parameters specifying how performance monitoring can be conducted by the UE. For an actual performance monitoring to be conducted, a selection of available (configured) configuration parameters may then be made in order to define how the actual performance monitoring is to be conducted. The performance monitoring configuration information may thus enable configurability of the UE with respect to performance monitoring tasks. Such configurability can be used to select from different possible configuration parameters, wherein respective performance monitoring tasks performed using such different parameters may provide performance measuring results for different monitoring items / sources, thereby facilitating the identification of root causes of performance degradation at such monitoring items / sources.

[0075] Once generated, the performance monitoring result may then be transmitted to the network node (step S1104). In case of UE-side performance monitoring, the performance monitoring result may correspond to a performance monitoring report generated by the UE. As mentioned above, such performance monitoring report may comprise a desired target KPI to be obtained by the performance monitoring (but may not comprise a target CSI or target CSI report,Telefonaktiebolaget LM Ericsson (publ) - 22 - 30A-169 639contrary to NW-side monitoring), and the transmission of the performance monitoring report to the network node may serve to inform the network node about the desired target KPI accordingly. In case of network-side performance monitoring, the performance monitoring result may comprise a target CSI report, allowing to calculate a desired target KPI at the network node, and the transmission of the target CSI report to the network node may serve to provide the network node with the necessary information to calculate the desired target KPI. The performance monitoring result transmitted from the UE to the network node may thus comprise either a target CSI report (for network-side performance monitoring) or a performance monitoring report generated by the UE (via UE-side performance monitoring). The desired target KPI may be, for example, an intermediate KPI (e.g., a normalized mean square error (NMSE), a squared generalized cosine similarity (SGCS), a generalized cosine similarity (GCS), or the like), and eventual KPI (e.g., a user perceived throughput (UPT), an expected block error rate (BLER), or the like) or a KPI indicative of a data drift either in an AI / ML input or an AI / ML output (or both) on the UE in case the UE runs at least one AI / ML model supporting the AI / ML enabled features described above.

[0076] Configuring the UE with the performance monitoring configuration information may be performed in accordance with (or "dependent on") certain UE capabilities associated with the performance monitoring. The method performed by the UE may thus further comprise transmitting, to the network node, UE capability information indicating at least one capability of the UE in relation to the performance monitoring. By such capability reports, the UE may inform the network node of its capabilities related to the performance monitoring described herein.

[0077] To determine a desired target KPI, different types of target CSI may be employed (i.e., can be reported by the UE to the network node for NW-side performance monitoring or can be used by the UE to conduct UE-side performance monitoring). The UE capability information transmitted to the network node may thus comprise one or more target CSI types supported by the UE for the performance monitoring or, in other words, the TSF compression performance monitoring capability report may include information on the supported target CSI. In particular, there may be two different types of target CSI, namely the target CSI that is obtained from the (uncompressed) predictive channels / CSI (such target CSI is referred to herein as "predicted CSI"), or the target CSI that is obtained from the measurement(s) of the CSI-RS(s) in the prediction window (such target CSI is referred to herein as "measured CSI"). The one or more target CSI types may thus include at least one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE. In other words, the TSF compression performance monitoring capabilityTelefonaktiebolaget LM Ericsson (publ) - 23 - 30A-169 639may be said to include one or more target CSI types (e.g., predicted-csi and measured-csi) that are supported by the UE for a performance monitoring operation. These two types of target CSI may entail different measurement / processing operations.

[0078] A 3GPP specification may not define separate UE capability parameters / information elements (IEs) for different life cycle management (LCM) operations (e.g., model inference, performance monitoring, data collection) of an AI / ML based feature. Whether a UE can support a certain LCM operation (e.g., performance monitoring) may instead be implicitly indicated by the fact whether the UE can support the required measurements and / or reporting operations needed for this LCM operation. The performance monitoring capability of a UE may thus be indicated by the UE's capability of reporting one or more target CSI formats (e.g., measured CSI in the prediction window and / or predicted CSI in the prediction window) in case of NW-side performance monitoring, or may be indicated by the UE's capability of reporting one or more performance monitoring metrics (e.g., desired target KPI, model accuracy, confidence level, etc.) in case of UE-side performance monitoring.

[0079] For Al based TSF compression, the CSI prediction part may be performed using Al based or non-AI based CSI prediction. The UE may support one or more CSI prediction mechanisms, wherein the complexity, processing time and / or processing resources needed to perform such prediction mechanisms may vary. This may also impact the performance monitoring operations that can be supported by the UE. Therefore, the UE may report the CSI prediction mechanisms that are supported by the UE in relation to a performance monitoring operation or, in other words, the UE capability report for TSF compression performance monitoring may include information on which CSI prediction performance monitoring is supported by the UE. For example, the type of target CSI that is supported by the UE for its performance monitoring operation may include at least one of measured-csi (that is obtained from the measurement of the CSI-RS in the prediction window), predicted-csi-non-ai and predicted-csi-ai. The UE capability information transmitted to the network node may thus comprise an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI, wherein the one or more CSI prediction mechanisms may include at least one of an Al based CSI prediction mechanism predicted-csi-ai) and a non-AI based CSI prediction mechanism predicted-csi-non-ai).

[0080] Performance monitoring operations may require a certain processing time, which may differ depending on the UE capability. Information on the processing time required by the UE to conduct performance monitoring operations may be needed by the network node for the purpose of an efficient performance monitoring configuration making sure that the UE is capable of finishing one performanceTelefonaktiebolaget LM Ericsson (publ) - 24 - 30A-169 639monitoring operation before the next operation / action (e.g., reporting the target CSI or the performance monitoring report) is performed by the UE. The UE capability report for TSF compression may thus include information on the time needed by the UE to finish the respective performance monitoring operations or, in other words, the UE capability report for TSF compression may include the time needed to conduct a TSF compression performance monitoring operation. The UE capability information transmitted to the network node may thus comprise an indication of a processing time required to complete a performance monitoring operation. The indication / information may be given in terms of a certain time units, such as seconds, milliseconds, slots, etc. The reported capability may be reported in a per subcarrier spacing (SCS) manner, for example.

[0081] As mentioned, the target CSI reported or used by the UE may vary. The information on the required processing time may thus be reported based on the target CSI type or, in other words, the indication of the processing time may be made per target CSI type. For example, the UE may report a first required processing time for measured CSZand a second processing time for predicted CSI. Similarly, if the UE supports both Al and non-AI based prediction, the UE may report at least one of a first required processing time for measured-csi, a second required processing time for predicted-csi-non-ai, and a third required processing time for predicted-csi-ai, for example.

[0082] The required processing time may also depend on the performance monitoring type (i.e., network-side performance monitoring or UE-side performance monitoring) or, in other words, dependent on which action the UE shall execute for its performance monitoring operation. For example, when the UE needs to report the measured CSI to the network node (i.e., NW side monitoring), the UE may only need to quantize the CSI-RS measurements (in the prediction window) before reporting it to the network node. On the other hand, when the UE needs to report a performance monitoring report (i.e., UE-side monitoring) (e.g., by comparing the compressed and uncompressed version of the CSI obtained from the CSI-RS measurement in the prediction window), the UE may need to further compress the measured CSI-RS, which most likely requires more time than simply quantizing the measurement. As such, the capability information on the required processing time may be reported per performance monitoring type (e.g., NW-side monitoring vs. UE-side monitoring I target CSI report vs. performance monitoring report). In other words, the indication of the processing time may be made per performance monitoring type, wherein the performance monitoring types may include at least one of network-side performance monitoring and UE-side performance monitoring, for example.Telefonaktiebolaget LM Ericsson (publ) - 25 - 30A-169 639

[0083] As described above, the UE may support different mechanisms for how TSF domain CSI compression is done, namely that CSI prediction and CSI compression is conducted in a joint (or "combined") block, or is conducted in separate blocks. The different types of mechanisms may result in different configurations and / or indications of performance monitoring that may be applied to the UE. For example, when the UE employs a joint prediction-compression model, the UE may not be configured to use or report uncompressed prediction channel / CSI as target CSI labels for performance monitoring, i.e., the target CSI may only be in terms of CSI based on the measurement of the CSI-RS in the prediction window. When the UE employs separate models for prediction and compression, on the other hand, the UE may be configured to use or report (uncompressed) predicted channels CSI as target CSI labels for performance monitoring. As another example, when the UE employs a joint prediction-compression model, the UE may not be configured to report a desired target KPI for prediction only, while, when the UE employs separate models for prediction and compression, the UE may be configured to report a desired target KPI associated with the CSI prediction model. Thus, the capability report of the UE may include information on how the prediction and compression are done by the UE (e.g., separate or joint) or, in other words, the UE capability information transmitted to the network node may comprise an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly.

[0084] In a variant, such capability information may also be inherited from the UE capability information related to the TSF compression feature itself. For example, a UE reporting a capability of supporting a separate prediction and compression feature may inherently have the capability of conducting performance monitoring for the separate prediction-compression mechanism. Similarly, a UE reporting a capability of supporting a joint prediction-compression feature may implicitly have the capability to support performance monitoring for the joint prediction-compression mechanism. Alternatively, such capability information may also be inherited from the Al model information.

[0085] For the case of separate CSI prediction-compression models, the UE may be able to produce (uncompressed) predicted channels / CSI before using it as the input for the compression block. The capability of conducting performance monitoring for the CSI prediction feature may thus be said to be a prerequisite for the performance monitoring capability of TSF compression or, in other words, the UE capability report of CSI prediction performance monitoring may be said to be a prerequisite for TSF compression performance monitoring. When the UE is thus configured to perform CSI prediction and CSI compression separately, the UE capability information transmitted to the network node may comprise an indication that the UE supports CSI predictionTelefonaktiebolaget LM Ericsson (publ) - 26 - 30A-169 639performance monitoring, wherein the indication that the UE supports CSI prediction performance monitoring may be a prerequisite for a capability of the UE for performing TSF compression. In certain variants, the capability of TSF compression performance monitoring may also be reported under the capability of conducting performance monitoring of CSI prediction or under the capability of supporting the CSI prediction feature, for example.

[0086] As said, the performance monitoring configuration information may be a dedicated configuration applied to the UE, wherein the configuration defines one or more parameters specifying how performance monitoring can be conducted by the UE. In other words, the UE may be configured with a performance monitoring configuration. Such configuration may be applied in a step prior to step SI 102, i.e., generating performance monitoring results in accordance with the performance monitoring configuration information.

[0087] As mentioned, there may be one or more types of target CSI that shall be reported (in case of NW-side monitoring) or be used (in case of UE-side monitoring) by the UE for its performance monitoring operations. Therefore, when it comes to configuring the UE, the UE may be configured with at least one target CSI type or, in other words, the performance monitoring configuration may include configuration on the target CSI that shall be reported by the UE (in case of NW-side monitoring) or shall be used by the UE for the performance monitoring report (in case of UE-side monitoring). The at least one parameter defined by the performance monitoring configuration information may as such comprise a target CSI type to be used for the performance monitoring. The target CSI type that can be configured for the UE may depend on the reported UE capability, e.g., according to the UE capability information comprising one or more target CSI types supported by the UE for the performance monitoring mentioned above. Accordingly, the target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring (measured-csi), and a predicted CSI obtained from CSI prediction performed by the UE predicted-csif As an example, such configuration may be as follows:performance-monitoring-configtarget-csi-type SEQUENCE (SIZE (1..2)) of {measured-csi, predicted-csi}

[0088] In a variant, the target CSI type may also differ based on the CSI prediction mechanism to be used by the UE for its performance monitoring operation. As an example, such configuration may be as follows:performance-monitoring-configTelefonaktiebolaget LM Ericsson (publ) - 27 - 30A-169 639target-csi-type SEQUENCE (SIZE (1..3)) of {measured-csi, predicted-csi-non-ai, predicted-csi-ai}Thus, in other words, when the target CSI type is a predicted CSI, the target CSI type may be one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms, wherein the underlying CSI prediction mechanisms may include at least an Al based CSI prediction mechanism and a non-AI based CSI prediction mechanism.

[0089] In a further variant, the target CSI type to be used by the UE for its performance monitoring operations may be configured per performance monitoring operation (e.g., target CSI report / NW-side monitoring vs. performance monitoring report / UE-side monitoring). As an example, such configuration may be as follows (for the case in which the UE is configured with measured-csi c\ predicted-csiy.performance-monitoring-configtarget-csi-reporttarget-csi-type SEQUENCE (SIZE (1..2)) of {measured-csi, predicted-csi} performance-monitoring-reporttarget-csi-type SEQUENCE (SIZE (1..2)) of {measured -csi, predicted-csi}Thus, in other words, in the performance monitoring configuration information, the target CSI type may be defined per performance monitoring type, wherein performance monitoring types may include at least one of network-side performance monitoring and UE-side performance monitoring. A similar example may apply for the case in which the UE is configured with measured-csi, predicted-csi-non-ai and predicted-csi-ai in its performance monitoring configuration of the target CSI.

[0090] As mentioned above, a certain target CSI type may entail a larger processing time or resource consumption for the UE (and possibly larger NW resources) than another target CSI type. In view of this, it may be beneficial that the UE is configured with a different periodicity for each target CSI type. The UE may thus be configured with a separate performance monitoring periodicity for each target CSI type or, in other words, a performance monitoring periodicity may be defined for each target CSI type in the performance monitoring configuration information. As an example, such configuration may be as follows:performance-monitoring-configmeasured-csi-periodicity INTEGER (Pl_min..Pl_max)predicted-csi-periodicity INTEGER (P2_min..P2_max)Telefonaktiebolaget LM Ericsson (publ) - 28 - 30A-169 639In this example, Pl_min, Pl_max, P2_min and P2_max may be the minimum periodicity of using measured CSI, the maximum periodicity of using measured CSI, the minimum periodicity of using predicted CSI, and the maximum periodicity of using predicted CSI, respectively, for performing monitoring operations. The value of the above parameter may be in time units, such as slots, seconds, milliseconds, etc. In a variant, a value of None may also be employed, indicating that performance monitoring with such a target CSI type is not needed.

[0091] In one implementation, applying the performance monitoring configuration to the UE may be performed after transmitting the UE capability information from the UE to the network node. Thus, in other words, the configuration may be applied depending on the reported UE capability. The UE may receive a corresponding monitoring configuration from the network node accordingly (e.g., upon the network node choosing a certain configuration depending on the received UE capabilities). The method performed by the UE may thus further comprise receiving the performance monitoring configuration information from the network node. This step may be performed in a step prior to step S1102, i.e., generating performance monitoring results in accordance with the performance monitoring configuration information.

[0092] It is to be noted that, even though the configuration examples provided above use the element performance-monitoring-config for the network node to configure performance monitoring related configurations in the UE, it does not necessarily mean that a new Radio Resource Control (RRC) Information Element (IE) must be defined in the specification for performance monitoring purposes. It will be understood that the configuration for performance monitoring could likewise be applied by reusing a legacy CSI reporting framework with corresponding modifications.

[0093] In another variant, the performance monitoring configuration information may be derived from other configurations (e.g., not explicitly signaled to the UE from the network node). In such a case, the method performed by the UE may further comprise deriving the performance monitoring configuration information from a configuration for a CSI- RS in a prediction window for performance monitoring. For example, the configuration of the target CSI type may be derived from the configuration on the CSI-RS in the prediction window. As an example, if the UE is configured with CSI-RS in the prediction window for performance monitoring, the UE may assume that the UE is configured with target CSI type measured-csi, whereas, if the UE is not configured with CSI-RS in the prediction window, the UE may assume that the UE is configured with target CSI type predicted-csi.

[0094] As said, the performance monitoring configuration information may be a dedicated configuration applied to the UE, wherein the configuration defines one orTelefonaktiebolaget LM Ericsson (publ) - 29 - 30A-169 639more parameters specifying how performance monitoring can be conducted by the UE. For an actual performance monitoring to be conducted, a selection of available / configured configuration parameters may then be made in order to define how the actual performance monitoring is to be conducted. Such selection of available / configured configuration parameters may be made via a separate message, such as a message providing an indication / instruction to the UE that performance monitoring is to be conducted (e.g., to start the performance monitoring operation). The method performed by the UE may thus further comprise receiving, from the network node, an indication to conduct performance monitoring, and such indication may contain information on which subset of configurations from the performance monitoring configuration applied to the UE is to be executed by the UE when performing performance monitoring. In such a variant, the indication may include the performance monitoring configuration information as a selection among available configuration options configured for the UE. The UE may start conducting the performance monitoring only after receiving such indication / instruction (or "command"), for example. As an example, the UE may receive an indication which target CSI type, among several supported target CSI types, shall be used for its performance monitoring operations at a particular time. The indication may be provided in the form of a Downlink Control Information (DCI) or a Media Access control (MAC) Control Element (CE), or the like. Such indication may deliver faster information to the UE, should a change on the target CSI type be needed.

[0095] It will be understood that the indication may also represent a mere instruction, i.e., without selection of a subset of the performance monitoring configurations, to conduct performance monitoring. In such a case, the UE could already be configured by a certain configuration, and the indication / instruction may then simply indicate to the UE that monitoring is now to be started. It will also be understood that the above-described indication / instruction is generally optional, because the UE might be configured to provide periodic performance monitoring results / reports by itself, for example, so that a trigger from the network node is not needed. In general, the UE may conduct performance monitoring operation in accordance with the indication, e.g., until a further indication is received (e.g., the further indication may specify a different selection of a subset of the performance monitoring configurations). The further indication may be another received indication (e.g., using DCI or MAC CE) or may trigger by the expiration of a timer or a counter, for example.

[0096] In one variant, the indication may be made via a bitfield contained in control information received by the UE from the network node, wherein the control information may comprise one of a DCI and a MAC CE, for example. In other words,Telefonaktiebolaget LM Ericsson (publ) - 30 - 30A-169 639the performance monitoring may be indicated by a bitfield in the DCI or a MAC CE. The bitfield may include an indication of one or more target CSI that shall be reported by the UE (in case of NW-side performance monitoring) or be used by the UE for the performance monitoring report (in case of UE-side performance monitoring). The size of the bitfield may be dependent on a number of target CSI types configured for the UE. For example, if the UE is configured with only one target CSI type, the size of the bitfield indication may be 0 bit, when the UE is configured with two or three target CSI, the bitfield for the indication may have a size of 1 or 2 bits, respectively. In general, the size of the bitfield may be formulated as ceil(log2(T)), where T is the number of target CSI types configured for the UE.

[0097] Regarding the interpretation of each bit's meaning in such bitfield, various implementations are conceivable. In one variant, bit interpretation may be defined in the text of a standard. For example, when two target CSI types are configured, a bit value of 0 may represent the predicted-csi, whereas a bit value of 1 may represent the measured-csi (or vice versa). In another variant, bit interpretation may depend on the NW configuration. For example, the UE may be configured with {measured-csi, predicted-csi} in its performance monitoring configuration. With such a configuration, a bit value of 0 may represent an indication to use measured-csi, whereas a value of 1 may represent an indication to use predicted-csi. Contrary, if the UE is configured with {predicted-csi, measured-csi } in its performance monitoring configuration, a bit value of 0 may represent an indication to use predicted-csi, whereas a value of 1 may represent an indication to use measured-csi. In a still further variant, a bitmap representation may be used for interpretation. For example, a first bit may represent an indication of whether the UE needs to use a first target CSI type, the second bit may represent an indication of whether the UE needs to use a second target CSI type, and so on. The bitmap representation may be defined in the text of a standard, or may be configurable by the NW, for example. Here, the UE may be configured to use more than one target CSI type. For example, the UE may receive an indication with a bit value of 11 and, thus, the UE needs to use both the first and the second target CSI type for its performance monitoring operations. To sum up, bits of the bitfield may be interpreted in accordance with at least one of a meaning defined in a standard, a network configuration, and a bitmap representation in which each bit is assigned a particular meaning.

[0098] Once a performance monitoring operation has started, a performance monitoring result may be generated in accordance with the performance monitoring configuration information, as described above. In one variant, when performing NW-side performance monitoring, the UE may be configured to report a target CSI to the network node. If the UE is configured to conduct performance monitoring withTelefonaktiebolaget LM Ericsson (publ) - 31 - 30A-169 639predicted CSI, the UE may measure the CSI-RS transmitted by the NW in the measurement window and then predict the CSI based on this measurement. This prediction result may then serve as the target CSI that is reported to the network node. If the UE is configured to conduct performance monitoring with measured CSI, the UE may measure the CSI-RS transmitted by the NW in the prediction window and use it as the target CSI that is reported to the network node.

[0099] In another variant, when performing UE-side performance monitoring, the UE may be configured to transmit the performance monitoring report / metric / output to the network node. If the UE is configured to use predicted CSI as the target CSI type, the UE may conduct a measurement to the CSI-RS in the measurement window and conduct CSI prediction. The result of this prediction may then be compressed using the CSI compression block. The output of the CSI compression block may then be decompressed with a nominal decoder (that can be obtained via model transfer, model training, standardized reference model, or the like). The output of the decompression block may then be compared to the predicted CSI obtained before to result in a desired target KPI (e.g., an intermediate KPI, such as SGCS). If the UE is configured to use measured CSI as the target CSI type, the UE may conduct a measurement to the CSI-RS in the measurement window and conduct CSI prediction. The result of this prediction may then be compressed using the CSI compression block. Besides measuring the CSI-RS in the measurement window, the UE may also measure the CSI-RS in the prediction window to obtain the measured CSI. The output of the CSI compression block may then be decompressed with a nominal decoder (that can be obtained via model transfer, model training, standardized reference model, or the like). The output of the decompression block may then be compared to the measured CSI to result in a desired target KPI (e.g., an intermediate KPI, such as SGCS). If the UE is configured to use both the measured CSI and the predicted CSI as the target CSI type, the UE may conduct two sets of performance monitoring report / output for this TSF compression feature, with one set corresponding to using the predicted CSI as target CSI, and the other set corresponding to using the measured CSI as target CSI.

[0100] Once the performance monitoring result is generated, it may be transmitted to the network node in accordance with step SI 104. When NW-side performance monitoring is used, the UE may transmit the target CSI to the network node, as described above, e.g., alongside the compressed predicted CSI. These two may be reported in a single container, or in a separate container. The container may be a CSI report, a MAC CE, or an RRC message, for example. The target CSI report in which the target CSI is transmitted may contain an indication of whether the target CSI sent by the UE is based on predicted-csi (CSI prediction outputs) and / or measured-Telefonaktiebolaget LM Ericsson (publ) - 32 - 30A-169 639csi (CSI obtained from the measurement of the CSI-RS in the prediction window). Thus, when the performance monitoring result comprises a target CSI report, the performance monitoring result may further comprise an indication of a target CSI type based on which the target CSI report is generated.

[0101] In one variant, the target CSI (which may consist of CSIs for multiple slots) may be transmitted as a single target CSI report. For example, the transmission may use a similar mechanism to the 3GPP Rel. 18 eType II codebook. To obtain better fidelity of the target CSI report, an enhancement of the parameters of the Rel. 18 eType II codebook may be used specifically for the target CSI reporting. For example, a larger value of quantization bits, number of beams, number of taps, or number of Doppler factors may be used.

[0102] In another variant, the target CSI may be transmitted as a multiple target CSI report. The number of the target CSI reports contained therein may depend on the number of CSI that shall be compressed by the CSI compression block. Here, the target CSI report may be quantized with e.g. a mechanism similar to 3GPP Rel. 16 eType II report, possibly with enhanced parameter values to guarantee the fidelity of the target CSI report. It is noted in this regard that the expression multiple target CSI in this description relates to the multiple CSI that shall be compressed by the UE and does not necessarily relate to the number of the CSI reporting that needs to be done by the UE, i.e., the multiple target CSI may be reported in a single CSI reporting.

[0103] As mentioned above, the UE may be configured or indicated to use more than one target CSI for each performance monitoring operation. Configured with such a configuration, the UE may need to report more than one target CSI type to the network node. In one variant, the target CSI report for multiple target CSI types may be reported in a single container. In an example, the reporting of the multiple target CSI types may contain only the target CSI without additional information. The determination of the start and the end of each target CSI report in that container may be known by the length of each target CSI report (which can be defined by the text of a standard or be included in the performance monitoring configurations, for example). The order of the target CSI type in the container may follow a definition of a standard or may be interpreted according to an NW configuration. For example, the first X bits may belong to the target CSI type with index 0 and the performance monitoring configurations, and so on. In another example, the UE may report an additional one or more bits in front of each target CSI reports to indicate which target CSI type the following report has. In such a variant, the performance monitoring result may comprise a plurality of target CSI reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated. In a still further variant,Telefonaktiebolaget LM Ericsson (publ) - 33 - 30A-169 639the target CSI report for multiple target CSI types may be reported in a separate container.

[0104] At the network node, the compressed predicted CSI may then be used by the network node as its decoder input. The output of the decoder may then be compared with the target CSI reported by the UE.

[0105] When UE-side performance monitoring is used, the UE may transmit performance monitoring reports to the network node, as described above. A performance monitoring report may be indicative of a performance metric, such as a desired target KPI (e.g., SGCS) and an indication whether the performance of the AI / ML model is acceptable, above a certain threshold, or the like. The performance monitoring report may be reported separately or inside the same container with the compressed predicted CSI. The performance monitoring report transmitted from the UE to the network node may contain an indication of whether the performance monitoring report sent by the UE is based on predicted-csi (CSI prediction outputs) and / or measured-csi (CSI obtained from the measurement of the CSI-RS in the prediction window). Thus, when the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result may further comprise an indication of a target CSI type based on which the performance monitoring report is generated.

[0106] In one variant, the performance monitoring report may comprise a bitfield that indicates which target CSI type the UE used to calculate the performance monitoring metrics / results / outputs. The size of the bitfield may depend on the number of target CSI types that are configured for the performance monitoring operations. As an example, the bitfield may have a size of 0, 1 or 2 bits when the UE is configured with one, two or three target CSI types, respectively, i.e., the size of the bitfield may be formulated as ceil (log 2(T)). When multiple performance monitoring reports are configured for the UE, the bitfield may be located in front of each performance monitoring report. In such a variant, the performance monitoring result may comprise a plurality of performance monitoring reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated. The target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

[0107] Figure 12 illustrates a method which may be performed by the network node according to the present disclosure. The network node supports performance monitoring for TSF domain CSI compression. In step S1202, the network node receives a performance monitoring result from a UE (e.g., the UE 900), wherein theTelefonaktiebolaget LM Ericsson (publ) - 34 - 30A-169 639performance monitoring result is generated in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring. The operation of the network node may as such be complementary to the operation of the UE described above, as such, aspects described above with regard to the operation of the network node may be applicable to the operation of the network node described in the following as well, and vice versa. Unnecessary repetitions are thus omitted in the following. The network node may correspond to the network node 1000 and the processing circuitry 1002 of the network node may be configured such that the network node is operable to carry out the method steps described herein with respect to the network node.

[0108] The performance monitoring result may comprise one of a target CSI report and a performance monitoring report generated by the UE. The method may further comprise receiving, from the UE, UE capability information indicating at least one capability of the UE in relation to the performance monitoring. The UE capability information may comprise one or more target CSI types supported by the UE for the performance monitoring. The one or more target CSI types may include at least one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

[0109] The UE capability information may comprise an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI. The one or more CSI prediction mechanisms may include at least one of an Artificial Intelligence (Al) based CSI prediction mechanism and a non-AI based CSI prediction mechanism. The UE capability information may comprise an indication of a processing time required to complete a performance monitoring operation. The indication of the processing time may be made per target CSI type. The indication of the processing time may be made per performance monitoring type. The UE capability information may comprise an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly. When the UE is configured to perform CSI prediction and CSI compression separately, the UE capability information may comprise an indication that the UE supports CSI prediction performance monitoring. The indication that the UE supports CSI prediction performance monitoring may be a prerequisite for a capability of the UE for performing TSF compression.

[0110] The at least one parameter defined by the performance monitoring configuration information may comprise a target CSI type to be used for the performance monitoring. The target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.Telefonaktiebolaget LM Ericsson (publ) - 35 - 30A-169 639When the target CSI type is a predicted CSI, the target CSI type may be one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms. The underlying CSI prediction mechanisms may include at least an Al based CSI prediction mechanism and a non-AI based CSI prediction mechanism.

[0111] In the performance monitoring configuration information, the target CSI type may be defined per performance monitoring type. A performance monitoring periodicity may be defined for each target CSI type in the performance monitoring configuration information. The method may further comprise transmitting the performance monitoring configuration information to the UE.

[0112] The method may further comprise transmitting, to the UE, an indication to conduct performance monitoring. The indication may include the performance monitoring configuration information as a selection among available configuration options configured for the UE. The indication may be made via a bitfield contained in control information transmitted from the network node to the UE. The control information may comprise one of a DCI and a MAC CE. A size of the bitfield may be dependent on a number of target CSI types configured for the UE. Bits of the bitfield may be interpreted in accordance with at least one of a meaning defined in a standard, a network configuration, and a bitmap representation in which each bit is assigned a particular meaning.

[0113] When the performance monitoring result comprises a target CSI report, the performance monitoring result may further comprise an indication of a target CSI type based on which the target CSI report is generated. The performance monitoring result may comprise a plurality of target CSI reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated. When the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result may further comprise an indication of a target CSI type based on which the performance monitoring report is generated. The performance monitoring result may comprise a plurality of performance monitoring reports and may further comprise indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated. The target CSI type may be one of a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

[0114] Figure 13 illustrates a signaling diagram of an exemplary interaction between the UE 900 and a network node 1000 according to the present disclosure. In step 1, the UE 900 may transmit UE capability information to the network node 1000,Telefonaktiebolaget LM Ericsson (publ) - 36 - 30A-169 639informing the network node 1000 of the UE's capabilities regarding performance monitoring, as described above. In step 2, the network node 1000 may transmit performance monitoring configuration information to the UE 900 in order to configure the UE 900 with configuration options for performance monitoring, as described above. In step 3, the network node 1000 may transmit an indication / instruction to the UE 900 to conduct performance monitoring, e.g., to start performance monitoring in accordance with the performance monitoring configuration information, as described above. The UE may then generate performance measuring results in accordance with the performance monitoring configuration and transmit such performance monitoring results, in step 4, to the network node 1000, as described above.

[0115] As has become apparent from the above, the present disclosure provides a technique for performing performance monitoring for TSF domain CSI compression. By (1) providing corresponding UE capability information, (2) configuring the UE with corresponding performance monitoring configuration information and (3) providing, in the performance monitoring results transmitted from the UE to the network node, indications of a target CSI type based on which the performance monitoring report was generated, the UE and the network node may have (or "gain") a same understanding of which performance monitoring approaches (e.g., measured-CSI-based or predicted-CSI-based) the UE is capable of, and which performance monitoring approaches shall be applied, or are currently reported by the UE. The performance monitoring configuration information generally enables configurability of the UE with respect to performance monitoring tasks to be conducted. Such configurability can be used to select from different possible configuration parameters, wherein the respective performance monitoring tasks performed using such different parameters may provide performance measuring results for different monitoring items / sources, wherein each such monitoring item / source may be a potential root cause for performance degradation. This enables the network node and / or the UE to conduct a root cause analysis if a degradation in performance is observed. Based thereon, corresponding LCM procedures may also be triggered accordingly.

[0116] It is noted that, while it has been described that the UE may be capable of implementing Al based TSF domain CSI compression, it will be understood that the present disclosure shall not be limited thereto. The present disclosure often specifically relates to Al based CSI compression as mechanism to compress the CSI. However, it will be understood that the technique presented herein may also be practiced with non-AI based CSI compression methods, if performance monitoring is required also for such compression mechanisms. Likewise, it is noted that, while, in the present disclosure, Al based CSI prediction is mostly assumed to be used in the CSI prediction block, the present disclosure shall not be limited thereto and non-AI based CSITelefonaktiebolaget LM Ericsson (publ) - 37 - 30A-169 639prediction may be employed in connection with the technique presented herein as well. Also, while many examples presented herein refer to a separate CSI prediction and CSI compression block, the present disclosure shall not be limited thereto and aspects of the present disclosure may likewise be practiced with a joint CSI prediction and CSI compression approach, wherein, in such an approach, the CSI prediction and the CSI compression may be performed by a single unified model. In such an approach, the model input may be the measurements (possibly pre-processed) of the CSI-RS in the measurement window and the output may be the compressed CSI of the predicted channels. Furthermore, it will be appreciated by those skilled in the art that the expression "predicted CSI" generally used herein may mean (or "cover") similar expressions, such as predicted channel, predicted PMI, predicted eigenvector, or the like. Moreover, although the term AI / ML model as referred to herein sometimes uses a single form, it will be understood that such AI / ML model may comprise more than one AI / ML model, e.g., in the form of AI / ML submodels. The UE may autonomously switch (or may be configured / instructed to switch) between the models depending on certain conditions and / or proprietary implementations, for example.

[0117] It is believed that the advantages of the technique presented herein will be fully understood from the foregoing description, and it will be apparent that various changes may be made in the form, constructions and arrangement of the exemplary aspects thereof without departing from the scope of the disclosure or without sacrificing all of its advantageous effects. Because the technique presented herein can be varied in many ways, it will be recognized that the present disclosure should be limited only by the scope of the embodiments that follow.

[0118] Group A Embodiments1. A method performed by a user equipment, UE, supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the method comprising:generating a performance monitoring result in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring; andtransmitting the generated performance monitoring result to a network node.2. The method of Embodiment 1, wherein the performance monitoring result comprises one of:a target CSI report, anda performance monitoring report generated by the UE.Telefonaktiebolaget LM Ericsson (publ) - 38 - 30A-169 6393. The method of Embodiment 1 or 2, further comprising:transmitting, to the network node, UE capability information indicating at least one capability of the UE in relation to the performance monitoring.4. The method of Embodiment 3, wherein the UE capability information comprises one or more target CSI types supported by the UE for the performance monitoring.5. The method of Embodiment 4, wherein the one or more target CSI types include at least one of:a measured CSI obtained by the UE from CSI measurement of a CSI-Reference Signal, CSI-RS, in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.6. The method of any one of Embodiments 3 to 5, wherein the UE capability information comprises an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI.7. The method of Embodiment 6, wherein the one or more CSI prediction mechanisms include at least one of:an Artificial Intelligence, Al, based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.8. The method of any one of Embodiments 3 to 7, wherein the UE capability information comprises an indication of a processing time required to complete a performance monitoring operation.9. The method of Embodiment 8, wherein the indication of the processing time is made per target CSI type.10. The method of Embodiment 8 or 9, wherein the indication of the processing time is made per performance monitoring type.11. The method of any one of Embodiments 3 to 10, wherein the UE capability information comprises an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly.Telefonaktiebolaget LM Ericsson (publ) - 39 - 30A-169 63912. The method of Embodiment 11, wherein, when the UE is configured to perform CSI prediction and CSI compression separately, the UE capability information comprises an indication that the UE supports CSI prediction performance monitoring.13. The method of Embodiment 12, wherein the indication that the UE supports CSI prediction performance monitoring is a prerequisite for a capability of the UE for performing TSF compression.14. The method of any one of Embodiments 1 to 13, wherein the at least one parameter defined by the performance monitoring configuration information comprises a target CSI type to be used for the performance monitoring.15. The method of Embodiment 14, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.16. The method of Embodiment 15, wherein, when the target CSI type is a predicted CSI, the target CSI type is one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms.17. The method of Embodiment 16, wherein the underlying CSI prediction mechanisms include at least:an Al based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.18. The method of any one of Embodiments 14 to 17, wherein, in the performance monitoring configuration information, the target CSI type is defined per performance monitoring type.19. The method of any one of Embodiments 14 to 18, wherein a performance monitoring periodicity is defined for each target CSI type in the performance monitoring configuration information.20. The method of any one of Embodiments 1 to 19, further comprising:receiving the performance monitoring configuration information from the network node.Telefonaktiebolaget LM Ericsson (publ) - 40 - 30A-169 63921. The method of any one of Embodiments 1 to 19, further comprising:deriving the performance monitoring configuration information from a configuration for a CSI-RS in a prediction window for performance monitoring.22. The method of any one of Embodiments 1 to 21, further comprising:receiving, from the network node, an indication to conduct performance monitoring.23. The method of Embodiment 22, wherein the indication includes the performance monitoring configuration information as a selection among available configuration options configured for the UE.24. The method of Embodiment 22 or 23, wherein the indication is made via a bitfield contained in control information received by the UE from the network node.25. The method of Embodiment 24, wherein the control information comprises one of:a Downlink Control Information, DCI, anda Media Access Control, MAC, Control Element, CE.26. The method of Embodiment 24 or 25, wherein a size of the bitfield is dependent on a number of target CSI types configured for the UE.27. The method of any one of Embodiments 24 to 26, wherein bits of the bitfield are interpreted in accordance with at least one of:a meaning defined in a standard,a network configuration, anda bitmap representation in which each bit is assigned a particular meaning.28. The method of any one of Embodiments 1 to 27, wherein, when the performance monitoring result comprises a target CSI report, the performance monitoring result further comprises an indication of a target CSI type based on which the target CSI report is generated.29. The method of any one of Embodiments 1 to 28, wherein the performance monitoring result comprises a plurality of target CSI reports and further comprisesTelefonaktiebolaget LM Ericsson (publ) - 41 - 30A-169 639indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated.30. The method of any one of Embodiments 1 to 29, wherein, when the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result further comprises an indication of a target CSI type based on which the performance monitoring report is generated.31. The method of any one of Embodiments 1 to 30, wherein the performance monitoring result comprises a plurality of performance monitoring reports and further comprises indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated.32. The method of any one of Embodiments 28 to 31, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.Group B Embodiments33. A method performed by a network node supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the method comprising:receiving a performance monitoring result from a user equipment, UE, wherein the performance monitoring result is generated in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring.34. The method of Embodiment 33, wherein the performance monitoring result comprises one of:a target CSI report, anda performance monitoring report generated by the UE.35. The method of Embodiment 33 or 34, further comprising:receiving, from the UE, UE capability information indicating at least one capability of the UE in relation to the performance monitoring.Telefonaktiebolaget LM Ericsson (publ) - 42 - 30A-169 63936. The method of Embodiment 35, wherein the UE capability information comprises one or more target CSI types supported by the UE for the performance monitoring.37. The method of Embodiment 36, wherein the one or more target CSI types include at least one of:a measured CSI obtained by the UE from CSI measurement of a CSI-Reference Signal, CSI-RS, in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.38. The method of any one of Embodiments 35 to 37, wherein the UE capability information comprises an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI.39. The method of Embodiment 38, wherein the one or more CSI prediction mechanisms include at least one of:an Artificial Intelligence, Al, based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.40. The method of any one of Embodiments 35 to 39, wherein the UE capability information comprises an indication of a processing time required to complete a performance monitoring operation.41. The method of Embodiment 40, wherein the indication of the processing time is made per target CSI type.42. The method of Embodiment 40 or 41, wherein the indication of the processing time is made per performance monitoring type.43. The method of any one of Embodiments 35 to 42, wherein the UE capability information comprises an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly.44. The method of Embodiment 43, wherein, when the UE is configured to perform CSI prediction and CSI compression separately, the UE capability information comprises an indication that the UE supports CSI prediction performance monitoring.Telefonaktiebolaget LM Ericsson (publ) - 43 - 30A-169 63945. The method of Embodiment 44, wherein the indication that the UE supports CSI prediction performance monitoring is a prerequisite for a capability of the UE for performing TSF compression.46. The method of any one of Embodiments 33 to 45, wherein the at least one parameter defined by the performance monitoring configuration information comprises a target CSI type to be used for the performance monitoring.47. The method of Embodiment 46, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.48. The method of Embodiment 47, wherein, when the target CSI type is a predicted CSI, the target CSI type is one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms.49. The method of Embodiment 48, wherein the underlying CSI prediction mechanisms include at least:an Al based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.50. The method of any one of Embodiments 46 to 49, wherein, in the performance monitoring configuration information, the target CSI type is defined per performance monitoring type.51. The method of any one of Embodiments 46 to 50, wherein a performance monitoring periodicity is defined for each target CSI type in the performance monitoring configuration information.52. The method of any one of Embodiments 33 to 51, further comprising:transmitting the performance monitoring configuration information to the UE.53. The method of any one of Embodiments 33 to 52, further comprising:transmitting, to the UE, an indication to conduct performance monitoring.Telefonaktiebolaget LM Ericsson (publ) - 44 - 30A-169 63954. The method of Embodiment 53, wherein the indication includes the performance monitoring configuration information as a selection among available configuration options configured for the UE.55. The method of Embodiment 53 or 54, wherein the indication is made via a bitfield contained in control information transmitted from the network node to the UE.56. The method of Embodiment 55, wherein the control information comprises one of:a Downlink Control Information, DCI, anda Media Access Control, MAC, Control Element, CE.57. The method of Embodiment 55 or 56, wherein a size of the bitfield is dependent on a number of target CSI types configured for the UE.58. The method of any one of Embodiments 55 to 57, wherein bits of the bitfield are interpreted in accordance with at least one of:a meaning defined in a standard,a network configuration, anda bitmap representation in which each bit is assigned a particular meaning.59. The method of any one of Embodiments 33 to 58, wherein, when the performance monitoring result comprises a target CSI report, the performance monitoring result further comprises an indication of a target CSI type based on which the target CSI report is generated.60. The method of any one of Embodiments 33 to 59, wherein the performance monitoring result comprises a plurality of target CSI reports and further comprises indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated.61. The method of any one of Embodiments 33 to 60, wherein, when the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result further comprises an indication of a target CSI type based on which the performance monitoring report is generated.Telefonaktiebolaget LM Ericsson (publ) - 45 - 30A-169 63962. The method of any one of Embodiments 33 to 61, wherein the performance monitoring result comprises a plurality of performance monitoring reports and further comprises indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated.63. The method of any one of Embodiments 59 to 62, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.Group C Embodiments64. A computer program product comprising program code portions for performing the method of any one of the Group A Embodiments when the computer program product is executed on a user equipment, UE.65. The computer program product of Embodiment 64, stored on a computer readable recording medium.66. A computer program product comprising program code portions for performing the method of any one of the Group B Embodiments when the computer program product is executed on a network node.67. The computer program product of Embodiment 66, stored on a computer readable recording medium.68. A user equipment, UE, supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the UE comprising:processing circuitry configured to perform any of the steps of any of the Group A Embodiments.69. A network node supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the network node comprising:Telefonaktiebolaget LM Ericsson (publ) - 46 - 30A-169639processing circuitry configured to perform any of the steps of any of the Group B Embodiments.70. A system comprising a user equipment, UE, according to Embodiment 68 and a network node according Embodiment 69.

Claims

Telefonaktiebolaget LM Ericsson (publ) - 47 - 30A-169 639CLAIMS1. A method performed by a user equipment, UE, supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the method comprising:generating a performance monitoring result in accordance with performance monitoring configuration information defining at least one parameter for the performance monitoring; andtransmitting the generated performance monitoring result to a network node.

2. The method of claim 1, wherein the performance monitoring result comprises one of:a target CSI report, anda performance monitoring report generated by the UE.

3. The method of claim 1 or 2, further comprising:transmitting, to the network node, UE capability information indicating at least one capability of the UE in relation to the performance monitoring.

4. The method of claim 3, wherein the UE capability information comprises one or more target CSI types supported by the UE for the performance monitoring.

5. The method of claim 4, wherein the one or more target CSI types include at least one of:a measured CSI obtained by the UE from CSI measurement of a CSI-Reference Signal, CSI-RS, in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

6. The method of any one of claims 3 to 5, wherein the UE capability information comprises an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI.

7. The method of claim 6, wherein the one or more CSI prediction mechanisms include at least one of:an Artificial Intelligence, Al, based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.Telefonaktiebolaget LM Ericsson (publ) - 48 - 30A-169 6398. The method of any one of claims 3 to 7, wherein the UE capability information comprises an indication of a processing time required to complete a performance monitoring operation.

9. The method of claim 8, wherein the indication of the processing time is made per target CSI type.

10. The method of claim 8 or 9, wherein the indication of the processing time is made per performance monitoring type.

11. The method of any one of claims 3 to 10, wherein the UE capability information comprises an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly.

12. The method of claim 11, wherein, when the UE is configured to perform CSI prediction and CSI compression separately, the UE capability information comprises an indication that the UE supports CSI prediction performance monitoring.

13. The method of claim 12, wherein the indication that the UE supports CSI prediction performance monitoring is a prerequisite for a capability of the UE for performing TSF compression.

14. The method of any one of claims 1 to 13, wherein the at least one parameter defined by the performance monitoring configuration information comprises a target CSI type to be used for the performance monitoring.

15. The method of claim 14, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.

16. The method of claim 15, wherein, when the target CSI type is a predicted CSI, the target CSI type is one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms.

17. The method of claim 16, wherein the underlying CSI prediction mechanisms include at least:Telefonaktiebolaget LM Ericsson (publ) - 49 - 30A-169 639an Al based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.

18. The method of any one of claims 14 to 17, wherein, in the performance monitoring configuration information, the target CSI type is defined per performance monitoring type.

19. The method of any one of claims 14 to 18, wherein a performance monitoring periodicity is defined for each target CSI type in the performance monitoring configuration information.

20. The method of any one of claims 1 to 19, further comprising:receiving the performance monitoring configuration information from the network node.

21. The method of any one of claims 1 to 19, further comprising:deriving the performance monitoring configuration information from a configuration for a CSI-RS in a prediction window for performance monitoring.

22. The method of any one of claims 1 to 21, further comprising:receiving, from the network node, an indication to conduct performance monitoring.

23. The method of claim 22, wherein the indication includes the performance monitoring configuration information as a selection among available configuration options configured for the UE.

24. The method of claim 22 or 23, wherein the indication is made via a bitfield contained in control information received by the UE from the network node.

25. The method of claim 24, wherein the control information comprises one of:a Downlink Control Information, DCI, anda Media Access Control, MAC, Control Element, CE.

26. The method of claim 24 or 25, wherein a size of the bitfield is dependent on a number of target CSI types configured for the UE.Telefonaktiebolaget LM Ericsson (publ) - 50 - 30A-169 63927. The method of any one of claims 24 to 26, wherein bits of the bitfield are interpreted in accordance with at least one of:a meaning defined in a standard,a network configuration, anda bitmap representation in which each bit is assigned a particular meaning.

28. The method of any one of claims 1 to 27, wherein, when the performance monitoring result comprises a target CSI report, the performance monitoring result further comprises an indication of a target CSI type based on which the target CSI report is generated.

29. The method of any one of claims 1 to 28, wherein the performance monitoring result comprises a plurality of target CSI reports and further comprises indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated.

30. The method of any one of claims 1 to 29, wherein, when the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result further comprises an indication of a target CSI type based on which the performance monitoring report is generated.

31. The method of any one of claims 1 to 30, wherein the performance monitoring result comprises a plurality of performance monitoring reports and further comprises indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated.

32. The method of any one of claims 28 to 31, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.

33. A method performed by a network node supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the method comprising:receiving a performance monitoring result from a user equipment, UE, wherein the performance monitoring result is generated in accordance with performanceTelefonaktiebolaget LM Ericsson (publ) - 51 - 30A-169 639monitoring configuration information defining at least one parameter for the performance monitoring.

34. The method of claim 33, wherein the performance monitoring result comprises one of:a target CSI report, anda performance monitoring report generated by the UE.

35. The method of claim 33 or 34, further comprising:receiving, from the UE, UE capability information indicating at least one capability of the UE in relation to the performance monitoring.

36. The method of claim 35, wherein the UE capability information comprises one or more target CSI types supported by the UE for the performance monitoring.

37. The method of claim 36, wherein the one or more target CSI types include at least one of:a measured CSI obtained by the UE from CSI measurement of a CSI-Reference Signal, CSI-RS, in a prediction window for performance monitoring, and a predicted CSI obtained from CSI prediction performed by the UE.

38. The method of any one of claims 35 to 37, wherein the UE capability information comprises an indication of one or more CSI prediction mechanisms supported by the UE to obtain the predicted CSI.

39. The method of claim 38, wherein the one or more CSI prediction mechanisms include at least one of:an Artificial Intelligence, Al, based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.

40. The method of any one of claims 35 to 39, wherein the UE capability information comprises an indication of a processing time required to complete a performance monitoring operation.

41. The method of claim 40, wherein the indication of the processing time is made per target CSI type.Telefonaktiebolaget LM Ericsson (publ) - 52 - 30A-169 63942. The method of claim 40 or 41, wherein the indication of the processing time is made per performance monitoring type.

43. The method of any one of claims 35 to 42, wherein the UE capability information comprises an indication on whether the UE is configured to perform CSI prediction and CSI compression separately or jointly.

44. The method of claim 43, wherein, when the UE is configured to perform CSI prediction and CSI compression separately, the UE capability information comprises an indication that the UE supports CSI prediction performance monitoring.

45. The method of claim 44, wherein the indication that the UE supports CSI prediction performance monitoring is a prerequisite for a capability of the UE for performing TSF compression.

46. The method of any one of claims 33 to 45, wherein the at least one parameter defined by the performance monitoring configuration information comprises a target CSI type to be used for the performance monitoring.

47. The method of claim 46, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.

48. The method of claim 47, wherein, when the target CSI type is a predicted CSI, the target CSI type is one of a plurality of predicted CSI types distinguished by their underlying CSI prediction mechanisms.

49. The method of claim 48, wherein the underlying CSI prediction mechanisms include at least:an Al based CSI prediction mechanism, anda non-AI based CSI prediction mechanism.

50. The method of any one of claims 46 to 49, wherein, in the performance monitoring configuration information, the target CSI type is defined per performance monitoring type.Telefonaktiebolaget LM Ericsson (publ) - 53 - 30A-169 63951. The method of any one of claims 46 to 50, wherein a performance monitoring periodicity is defined for each target CSI type in the performance monitoring configuration information.

52. The method of any one of claims 33 to 51, further comprising:transmitting the performance monitoring configuration information to the UE.

53. The method of any one of claims 33 to 52, further comprising:transmitting, to the UE, an indication to conduct performance monitoring.

54. The method of claim 53, wherein the indication includes the performance monitoring configuration information as a selection among available configuration options configured for the UE.

55. The method of claim 53 or 54, wherein the indication is made via a bitfield contained in control information transmitted from the network node to the UE.

56. The method of claim 55, wherein the control information comprises one of:a Downlink Control Information, DCI, anda Media Access Control, MAC, Control Element, CE.

57. The method of claim 55 or 56, wherein a size of the bitfield is dependent on a number of target CSI types configured for the UE.

58. The method of any one of claims 55 to 57, wherein bits of the bitfield are interpreted in accordance with at least one of:a meaning defined in a standard,a network configuration, anda bitmap representation in which each bit is assigned a particular meaning.

59. The method of any one of claims 33 to 58, wherein, when the performance monitoring result comprises a target CSI report, the performance monitoring result further comprises an indication of a target CSI type based on which the target CSI report is generated.

60. The method of any one of claims 33 to 59, wherein the performance monitoring result comprises a plurality of target CSI reports and further comprisesTelefonaktiebolaget LM Ericsson (publ) - 54 - 30A-169 639indications respectively indicating a target CSI type for each of the plurality of target CSI reports based on which the respective target CSI report is generated.

61. The method of any one of claims 33 to 60, wherein, when the performance monitoring result comprises a performance monitoring report generated by the UE, the performance monitoring result further comprises an indication of a target CSI type based on which the performance monitoring report is generated.

62. The method of any one of claims 33 to 61, wherein the performance monitoring result comprises a plurality of performance monitoring reports and further comprises indications respectively indicating a target CSI type for each of the plurality of performance monitoring reports based on which the respective performance monitoring report is generated.

63. The method of any one of claims 59 to 62, wherein the target CSI type is one of:a measured CSI obtained by the UE from CSI measurement of a CSI-RS in a prediction window for performance monitoring, anda predicted CSI obtained from CSI prediction performed by the UE.

64. A computer program product comprising program code portions for performing the method of any one of claims 1 to 32 when the computer program product is executed on a user equipment, UE.

65. The computer program product of claim 64, stored on a computer readable recording medium.

66. A computer program product comprising program code portions for performing the method of any one of claims 33 to 63 when the computer program product is executed on a network node.

67. The computer program product of claim 66, stored on a computer readable recording medium.

68. A user equipment, UE, supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the UE comprising:Telefonaktiebolaget LM Ericsson (publ) - 55 - 30A-169 639processing circuitry configured to perform any of the steps of any of claims 1 to 32.

69. A network node supporting performance monitoring for Temporal-Spatial-Frequency, TSF, domain Channel State Information, CSI, compression, the network node comprising:processing circuitry configured to perform any of the steps of any of claims 33 to 63.

70. A system comprising a user equipment, UE, according to claim 68 and a network node according claim 69.

Images