Mechanism for controlling communication state management process

Abstract

An apparatus comprising at least one processor, and at least one memory for storing instructions that, when executed by the at least one processor, cause the apparatus at least to: determine an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, determine whether a modification of the configuration of the prediction functionality is required, based at least in part on the determination of the input change event, and in response to a determination that the modification is required, instruct the user equipment to modify the configuration of the prediction functionality.

Description

[0001] MECHANISM FOR CONTROLLING COMMUNICATION STATE MANAGEMENT PROCESS

[0002] DESCRIPTION

[0003] BACKGROUND

[0004] Field

[0005] Examples of embodiments described herein relate to apparatuses, methods, systems, computer programs, computer program products and (non-transitory) computer-readable media usable for controlling a communication state management process in a communication network. Specifically, examples of embodiments described herein relate to apparatuses, methods, systems, computer programs, computer program products and (non-transitory) computer-readable media usable for considering specific events in a communication state management process, such as a beam management (BM) process, using communication state prediction processing like beam prediction processing using, for example, Artificial Intelligence (Al) / Machine learning (ML) based models.

[0006] Background

[0007] The following description of background may include insights, discoveries, understandings or disclosures, or associations, together with disclosures that are not already known, but rather provided herein by the disclosure as one or more examples of embodiments. Some of examples of embodiments may be specifically pointed out below, whereas other of such contributions will be apparent from the related context.

[0008] SUMMARY

[0009] According to an example of an embodiment, there is provided, for example, an apparatus comprising at least one processor, and at least one memory for storing instructions that, when executed by the at least one processor, cause the apparatus at least to: determine an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, determine whether a modification of a configuration of the prediction functionality is required, based at least in part on the determination of the input change event, and in response to a determination that the modification is required, instruct the user equipment to modify the configuration of the prediction functionality. Furthermore, according to an example of an embodiment, there is provided, for example, a method comprising determining an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, determining whether a modification of a configuration of the prediction functionality is required, based at least in part on the determination of the input change event, and in response to a determination that the modification is required, instructing the user equipment to modify the configuration of the prediction functionality.

[0010] According to further refinements, these examples may include the following features:

[0011] - a determination of the input change event may comprise one of the following: to detect or predict occurrence of the input change event, or to receive an event notification from the user equipment, wherein to detect or predict the occurrence of the input change event may be based on at least one of the following: sensor data obtained from at least one sensor, wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value, or radio data obtained for the user equipment, wherein the input change event is detected or predicted by comparing values measured in a communication with the user equipment with a preset communication threshold value and a time threshold value; and the event notification received from the user equipment may comprise information indicating that the user equipment has detected or predicted the occurrence of the input change event;

[0012] - the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained may comprise at least one of the following: a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data may comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power;

[0013] - the event notification received from the user equipment may comprise at least one of the following: an identification of a type of the input change event, time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window, an indication of communication resources associated with one or more communication paths affected by the event, or an indication of an expected drop of communication quality caused by the event;

[0014] - the input change event may comprise one of the following: a a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment;

[0015] - the prediction functionality may be for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment;

[0016] - a configuration of the user equipment for triggering the input change event may be executed, and information for detecting or predicting the input change event and for preparing the event notification at the user equipment side may be provided;

[0017] - channel state information prediction reports and performance monitoring reports may be received from the user equipment, and at least one of the received channel state information prediction reports and performance monitoring reports may be used for determining whether a modification of the configuration of the prediction functionality is required;

[0018] - the determination whether the modification of the configuration of the prediction functionality is required may be based on at least one of the following: information about a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; information about communication resources associated with one or more communication paths affected by the event; information about an expected drop of communication quality caused by the event, or at least one of performance monitoring report indicating a communication state prediction performance on the user equipment side;

[0019] - the modification of the configuration of the prediction functionality may comprise one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting, an activation or deactivation of a machine learning based communication state prediction model, or a switching to a different machine learning based communication state prediction model;

[0020] - after it is determined that a modification of the configuration of the prediction functionality is required, a preset time interval may be waited, and the user equipment may be instructed to modify the configuration of the prediction functionality after the waiting time interval has expired;

[0021] According to an example of an embodiment, there is provided, for example, an apparatus comprising at least one processor, and at least one memory for storing instructions that, when executed by the at least one processor, cause the apparatus at least to: detect or predict an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, wherein the occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data received from at least one sensor, or radio data measured at the user equipment; in response to that the input change event is detected or predicted, prepare an event notification, and send, to a communication network, the prepared event notification. Furthermore, according to an example of an embodiment, there is provided, for example, a method comprising detecting or predicting an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, wherein the occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data received from at least one sensor, or radio data measured at the user equipment; in response to that the input change event is detected or predicted, preparing an event notification, and sending, to a communication network, the prepared event notification.

[0022] According to further refinements, these examples may include the following features:

[0023] - the input change event may be detected or predicted by comparing a preset sensor threshold value with a sensor output value, or the input change event may be detected or predicted by comparing values measured for a communication with the communication network with a preset communication threshold value and a time threshold value;

[0024] - the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained may comprise at least one of the following: a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data may comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power;

[0025] - the event notification sent to the communication network may comprise at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event;

[0026] - the input change event may comprise one of the following: a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment;

[0027] - the prediction functionality may be for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment;

[0028] - configuration information may be received from the communication network for triggering the input change event, and information for detecting or predicting the input change event and for preparing the event notification may be received; - channel state information prediction reports and performance monitoring reports related to the prediction functionality may be transmitted to the communication network;

[0029] - an instruction for modifying the configuration of the prediction functionality may be received, and the modification of the configuration of the prediction functionality may be executed, wherein the modification of the prediction functionality may comprise one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting, an activation or deactivation of a machine learning based communication state prediction model, or a switching to a different machine learning based communication state prediction model;

[0030] - the above defined apparatus and processing may be comprised in the user equipment or attachable to the user equipment.

[0031] In addition, according to embodiments, there is provided, for example, a computer program product for a computer, including software code portions for performing the steps of the above defined methods, when said product is run on the computer. The computer program product may include a computer-readable medium on which said software code portions are stored. Furthermore, the computer program product may be directly loadable into the internal memory of the computer and / or transmittable via a network by means of: upload; download; and / or push procedures.

[0032] BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Some examples of disclosure related to embodiments are described below, by way of example only, with reference to the accompanying drawings, in which:

[0034] Fig. 1 shows a diagram illustrating a communication network part where a beam management processing as an example for a communication state management processing is implemented;

[0035] Fig. 2 shows an example of a framework for a UE sided AI / ML model for beam management ;

[0036] Fig. 3 shows a diagram illustrating a comparison between predicted and actual beam power values in the presence of a blockage;

[0037] Fig. 4 shows a diagram illustrating a variation of error in predicted beam’s received power compared to strongest beam's received power; Fig. 5 shows a diagram illustrating a blockage gain;

[0038] Fig. 6 shows a signaling diagram illustrating an example of a processing according to an example of embodiments;

[0039] Fig. 7 shows a signaling diagram illustrating an example of a processing according to an example of embodiments;

[0040] Fig. 8 shows a signaling diagram illustrating an example of a processing according to an example of embodiments;

[0041] Fig. 9 shows a signaling diagram illustrating an example of a processing according to an example of embodiments;

[0042] Fig. 10 shows a flow chart of a processing conducted in a communication network control element or communication network control function conducting a processing according to some examples of embodiments;

[0043] Fig. 11 shows a flow chart of a processing conducted in a user equipment conducting a processing according to some examples of embodiments;

[0044] Fig. 12 shows a diagram of a communication network control element according to some examples of embodiments; and

[0045] Fig. 13 shows a diagram of a user equipment according to some examples of embodiments.

[0046] DETAILED DESCRIPTION

[0047] With the advancement of modern communication networks, such as 3GPP 5G or future 6G networks, highly directional beamforming architecture with a larger number of antenna elements is employed for achieving higher data rate. Special procedures are essential for the UE to establish connection and maintain the link even in mobility scenarios.

[0048] Beam management (BM) is a set of Layer 1 (PHY (physical)) and Layer 2 (MAC (Medium Access Control) procedures to establish and retain an optimal beam pair for good connectivity. A beam pair consists of a transmit (TX) beam and a corresponding receive (RX) beam in one link direction. Fig. 1 illustrates a simplified network architecture where a user equipment (UE) 10 is connected to a communication network, i.e. to a communication network control element or communication network control function, such as a gNB, 20 by using at least one of a plurality of communication beams 25 between the communication network 20 and the UE 10.

[0049] It is to be noted that even though in the following description reference is made to a user equipment (UE), examples of embodiments are applicable to any sort of user device, terminal device, Internet-of-Things user equipment and the like, or to a corresponding electronic device mounted to various other devices, such as vehicles, which shall be also meant by the term “UE”.

[0050] In communication networks, such as 3GPP based networks, for example, in a legacy BM process, the gNodeB (gNB) configures UEs to measure a set of Reference Signals (RSs) such as Synchronization Signal Block (SSB) and Channel State Information (CSI) RSs. The gNB transmits downlink (DL) transmit (TX) beams using unique RSs, and UE measures the reference signal received power (RSRP) of each beam with receiver (RX) beams and reports the strongest set of RS identifiers and corresponding RSRP measurements to gNB. The current CSI reporting may be configured in a periodic / semi-persistent fashion with a high periodicity in order to get a full and accurate picture of the UE conditions and optimize e.g., beam switching. Thus, a large number of beams included in TX and RX codebooks leads to perform a large number of measurements for determining the best TX and RX beam pair.

[0051] Hence, the legacy BM process is time consuming and not scalable when the size of antenna arrays increases. Therefore, Artificial Intelligence (Al) and Machine Learning (ML) algorithms are implemented so as to replace sequential beam scanning by recommending a reduced set of beams likely to contain the best beam index of the full beam scan.

[0052] In 3GPP, studies have been made regarding the support of an AI / ML-enabled radio interface for cellular systems. BM is one example for applying the AI / ML model in the New Radio (NR) radio interface. It is an aim to identify a common AI / ML framework, including functional requirements of AI / ML architecture, which could be used in subsequent releases and will pave the way for the design of native AI / ML networks, such as in 6G. The motivation behind AI / ML usage is an increasing demand for efficient beam prediction schemes in wireless communication systems. One goal is, for example, to demonstrate the effectiveness of AI / ML- based BM models in reducing measurement overhead (tightly related to the UE power consumption) while maintaining comparable end-to-end performance to conventional BM procedures. In the context of BM procedures, spatial-domain and time-domain beam prediction is considered. The scope of spatial-domain beam prediction (also referred to as BM-Case1) is to predict the best DL TX beam or DL TX / RX beam pairs in different spatial locations. On the other hand, time-domain beam predictions (also referred to as BM-Case2) aim to predict the best DL TX beam or DL TX / RX beam pairs to use for future time instants. It has been shown that ML algorithms enable predicting the serving beam for different UE locations and time instances, thus avoid measuring the actual beam quality and saving those resources for data transmission and for the UE to increase the length of Discontinuous Reception (DRX).

[0053] Basically, AI / ML models discussed in BM-Case1 and BM-Case2 can be trained and deployed, for example, at the gNB (i.e. NW-sided models) or the UE (i.e. UE-sided models). In Fig. 2, an example of a UE sided AI / ML model operation is explained.

[0054] In the training and deployment of the AI / ML model shown in Fig. 2, which shows the framework for a UE sided AI / ML model for beam management, data collection for model training may reuse existing beam measurement and reporting frameworks. For instance, the gNB configures the UE with a CSI reporting configuration to perform measurements of Set A (set of beams targeted for prediction) and Set B (set of beams used for measurement) beams as shown in S210.

[0055] In S220, the measurements are collected at the UE and at various time occasions and used for model training. The UE measures RS resources corresponding to different TX beams using an optimal RX beam determined by previous measurements. Also, the UE may filter the instantaneous Reference Signal Receiving Power (RSRP) measurements to mitigate the effects of fast fading with a Layer-1 (L1) filter and collect L1-RSRP measurements.

[0056] Once a trained AI / ML model is deployed at the UE, the gNB configures the UE in S230 to measure Set B beams to be used as input to the model during inference in S240. The UE predicts best K beams in Set A based, for example, on L1-RSRP measurements associated with Set B and then reports the predicted best K beams to the gNB in S250.

[0057] It is to be noted that the gNB, after the UE performs inference in S240 and reports in S250 the predicted best K beams to the gNB, may optionally perform an additional step of refined measurements of the bets K predicted beams. Alternatively, the predicted best K beam can be used directly without a step of refined measurements to make informed decisions about transmission control information (TCI) state activation and indication to the UE. In S260, the gNB send a DL control message to indicate the DL TX beam to be used for receiving DL channels.

[0058] In S270, the gNB sends to the UE Set A and Set B beams for performance monitoring (in S280). Performance monitoring is required for various reasons such as sudden changes in channel conditions or other changes to the NW or UE configurations, where the model may not always provide accurate predictions. If the AI / ML model performance is not satisfactory, AI / ML model may need to be deactivated, causing a fallback to the legacy BM process until AI / ML model able to predict the best beam with the expected accuracy.

[0059] That is, in addition to AI / ML model training and inference, a mandatory principle of AI / ML enabled BM is to ensure the performance of AI / ML model remains at a satisfactory level with respect to non-AI / ML legacy BM operations. For this, performance monitoring (PM) of AI / ML models and reporting of results of PM is conducted which allows to ensure that the model is working as expected.

[0060] Basically, PM is performed less frequently compared to AI / ML model inference events to reduce the burden on UL / DL signaling. There are different key performance indicators (KPI) conceivable which could be used for activating / deactivating the model. For example, for evaluating a real-time performance of an AI / ML model, beam prediction accuracy related KPIs, such as the prediction accuracy for the best I best K beams, link quality related KPIs. Such as throughput, L1-RSRP etc. of the predicted beam, a performance metric based on input / output data distribution of AI / ML, or an L1-RSRP difference evaluated by comparing measured RSRP and predicted RSRP can be implemented.

[0061] In the latter example, i.e. performance evaluation based on L1-RSRP error, the KPI used to AI / ML model activation / deactivation is based, for example, on a threshold (e) defined for L1- RSRP error where AI / ML model is deactivated and legacy BM is activated if L1-RSRP error greater than e. Otherwise, if L1-RSRP error less than e, the AI / ML model-predictions are maintained or can be re-activated. In essence, the AI / ML model predictions are assumed to be reliable if L1-RSRP error less than e to keep using the AI / ML model.

[0062] It has been found out that in AI / ML-based BM applications the AI / ML models are sensible to changes of input distributions, such as RSRP. One exemplary scenario is for instance when dynamic blockers are present in the environment and obstruct the beam(s) transmitted by the network (NW) and / or used by the UE to receive the signal. Therefore, it is expected that the AI / ML model performance may rapidly change in the presence of a blocker. In order to illustrate this, Fig. 3 shows a diagram illustrating a comparison between power values of predicted and actual TX beams over time in the presence of a blockage. Specifically, Fig. 3 shows the variations of RSRP values corresponding to the strongest TX beam when a blocker intersects the signal path between NW and UE, wherein the solid curve illustrates the values of the actually strongest TX beam, and the dotted-line curve illustrates the values for the predicted TX beam. As can be seen in Fig. 3, it is common to see a considerable loss in RSRP even for the strongest TX beam in case of blockage (indicated by dashed lines). Outside the blockage, the AI / ML model shows satisfactory prediction accuracy, i.e. a low RSRP error for a non-blocked UE. However, the prediction accuracy drops in the presence of a blocker wherein RSRP can be further penalized due to erroneous predictions. The reason for this is that models are generally trained in situation without the blockage, and when one or more channel paths become blocked, the distribution of the measurements changes determining a change of condition with respect to the those assumed during the training (i.e. without blockage). Thus, therefore the model is not capable of recognizing the right input-output relationship and it provides non-accurate prediction as output.

[0063] To analyze the above behavior, simulations have been conducted for a scenario where a blocker intersects the channel of a UE configured to perform AI / ML-based BM. The performance of the AI / ML model is measured with RSRP error epspp which is calculated as the difference between the predicted beam and the strongest ground-truth beam.

[0064] In Fig. 4, a diagram illustrating a variation of error in predicted beam’s received power compared to strongest beam's received power is shown, i.e. the variation of eRSRp with and without the blocker. It is evident that the RSRP error epspp becomes higher during a blockage event (indicated by the dashed box) compared with the non-blocked situation.

[0065] As indicated above, PM is used to determine, for example, activation / deactivation of an AI / ML model. Some predefined intervals or threshold criteria depending on the performance monitoring (PM) metrics trends overtime can be used for such purpose. In one example, AI / ML model activation deactivation is tested based on an RSRP error threshold applied the PM metric. In this example, the following criteria is used:

[0066] 6RSRP < 1 dB

[0067] PPM _ — I1’

[0068] 6RSRP — 1 dB where the AI / ML model is deactivated and fallback to legacy BM is made if pPM= 0 while the AI / ML model is activated or kept used if pPM= 1 . If the AI / ML model is deactivated, performance is calculated offline with the measurement reports received by the legacy BM. If pPM= 1 is observed, AI / ML model will be re-activated.

[0069] As indicated above, PM is executed less frequently than model inference. For example, for the considerations of the simulation, the PM event is performed once per 10 inference / beam prediction events. This is indicated in Fig. 4 by circles denoting the epspp observed at a time of PM-event.

[0070] As can be deduced from the results illustrated in Fig. 4, PM-based AI / ML model activation performs well in non-blocked situations. On the other hand, the PM-based AI / ML model activation is not very effective during a blockage. This is because, even though the AI / ML model is trained with blockers in the environment, the AI / ML model performances changes between extremes (from very accurate (low dB value) to extremely inaccurate (high dB value) in a small amount of time) for the whole-time duration when a blocker obstructs the link with the network.

[0071] That is, in addition to significant prediction errors, PM-based AI / ML model activation / deactivation has the following challenges in case of an event where the input distribution (i.e. received signal power, for example) changes, e.g. in the presence of a blocker.

[0072] As PM events are less frequent compared to the AI / ML model inference events, NW may take a long duration to observe higher epspp. and deactivate the AI / ML model. Thus, errors caused by the usage of the AI / ML model may be corrected late.

[0073] Furthermore, as depicted in Fig. 4, the AI / ML model may predict the best beam accurately from time to time. If the PM-event overlaps with such predictions, poor performance of the AI / ML model may not be captured in PM events.

[0074] Moreover, for example, during blocked time-instants, the PM-based AI / ML model activation / deactivation may lead to frequent AI / ML model activations and deactivations. This occurs, for example, at situations where a strong difference between the 6RSRP values can be observed between adjacent PM events. This imposes additional burden to signalling between NW and UE as each activation / deactivation requires configuration / reconfiguration work (that is, signalling overhead and related processing function is executed) at the NW and the UE, e.g. with different CSI report, and / or in certain cases to transmit RRC reconfiguration / configuration.

[0075] In certain scenarios, where the probability of, for example, blockage can be considered low, each PM instance shall not trigger AI / ML model switching, but multiple instances of PM should be used prior taking switching decisions. On the other hand, if there is higher probability that blockage occurs often in the network and they occur for longer time durations, this averaging of PM instances can impact the performance of the predictions. Therefore, a different configuration where each blocking scenario may trigger an ML model switch should be applied.

[0076] In order to deal with the above described issues, it is required that the NW is aware of the actual situation regarding changes affecting the performance of a prediction functionality related to predict a communication state, such as a beam to be used, which is caused, for example, by blockage. That is, the NW should be able to determine, e.g. by detection or information, such events (i.e. input change events), so that the NW can properly control whether or not a result of a (AI / ML based) communication state prediction procedure is accurate or not and thus to be used or provided, or not. For example, the ML model activation / deactivation or fallback to legacy procedure is target of such a control. For this, the NW requires event information impacting the ML model operation.

[0077] It is to be noted that in the following, as an example for a communication state prediction, a beam prediction is considered. However, the principles described below are applicable to various communication state prediction procedures using a prediction functionality, such as a prediction of channel state information, a prediction of a received signal strength, and the like; that is, a communication state which is the target of a prediction functionality to which examples of embodiments are related means various parts of a communication structure which can be predicted by using an AI / ML model or the like.

[0078] According to examples of embodiments, measures are proposed which allow the NW and the UE to address the above discussed issues. Specifically, according to examples of embodiments, there is proposed a mechanism by means of which an event-based AI / ML model activation / deactivation framework can be implemented.

[0079] In the following, different examples of embodiments will be described for illustrating several examples for processing applicable in the above indicated situations. It is to be noted that the following examples and embodiments are to be understood only as illustrative examples. Although the text herein may refer to “an”, “one”, or “some” example(s) or embodiment(s) in several locations, this does not necessarily mean that each such reference is related to the same example(s) or embodiment(s), or that the feature only applies to a single example or embodiment. Single features of different embodiments may also be combined to provide other embodiments. Furthermore, terms like “comprising” and “including” should be understood as not limiting the described embodiments to consist of only those features that have been mentioned; such examples and embodiments may also comprise features, structures, units, modules etc., that have not been specifically mentioned.

[0080] The general functions and interconnections of described elements and functions, which also depend on the system type, are understood by those skilled in the art and described in corresponding specifications so that a detailed description thereof may be omitted herein. However, it is to be noted that several additional elements and / or signaling links may be employed for a data exchange between involved elements, functions or applications, like a communication endpoint, a network control element, such as a server, a gateway, and other elements besides those described in detail herein below.

[0081] A system architecture as being considered in examples of embodiments may also be able to support the usage of cloud services for virtual network elements or functions thereof, wherein it is to be noted that the virtual system part can also be provided by non-cloud resources, e.g. an internal network or the like. Generally, a system element, or node, can be implemented either as a network element on a dedicated hardware, as a software instance running on a dedicated hardware, or as a virtualized function instantiated on an appropriate platform, e.g., a cloud infrastructure.

[0082] Furthermore, a network element or network function, such as network node acting as a communication network control element or communication network control function, like a gNB, or a user equipment (UE), as described herein, may be implemented by software, e.g. by a computer program product for a computer, and / or by hardware. For executing their respective processing, correspondingly used devices, nodes, functions or system elements may include several means, modules, units, components, etc. (not shown) which are utilized for control of processing and / or signaling functionality. Such means, modules, units and components may include, for example, one or more processors or processor units including one or more processing portions for executing instructions and / or programs and / or for processing data, storage or memory units or means for storing instructions, programs and / or data, for serving as a work area of the processor or processing portion and the like (e.g. read only memory (ROM), random access memory (RAM), electrically erasable programmable read only memory (EEPROM), and the like), input or interface means for inputting data and instructions by software (e.g. floppy disc, compact disc ROM (CD-ROM), EEPROM, and the like), a user interface for providing monitor and manipulation possibilities to a user (e.g. a screen, a keyboard and the like), other interface or means for establishing links and / or connections under the control of the processor unit or portion (e.g. wired and wireless interface means, radio interface means including e.g. an antenna unit or the like, means for forming a radio communication part etc.) and the like, wherein respective means forming an interface, such as a radio communication part, can be also located on a remote site (e.g. a radio head or a radio station etc.). It is to be noted herein that processing portions should not be only considered to represent physical portions of one or more processors, but may also be considered as a logical division of the referred processing tasks performed by one or more processors.

[0083] As used in this application, the term “circuitry” may refer to one or more or all of the following:

[0084] (a) hardware-only circuit implementations (such as implementations in only analog and / or digital circuitry) and

[0085] (b) combinations of hardware circuits and software, such as (as applicable):

[0086] (i) a combination of analog and / or digital hardware circuit(s) with software / firmware and

[0087] (ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a device acting as a node, to perform various functions) and hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation. This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used herein, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a system element or node or a similar integrated circuit in server, a communication network device, or other computing or network device.

[0088] Furthermore, as used herein, “at least one of the following:” and “at least one of ” and similar wording, where the list of two or more elements are joined by “and” or “or”, mean at least any one of the elements, or at least any two or more of the elements, or at least all the elements. According to examples of embodiments, there is provided a framework for controlling a communication state prediction procedure, for example a beam prediction procedure wherein specific events (referred to as input change events) related to a change of power in a received signal at a UE, e.g. an event related to a blockage of a communication path between the UE and the network, are determined in a suitable manner so that the NW is aware of this. A detection or prediction of the event as such can be executed on the UE side or the network side, wherein the UE is preferably configured by the NW for conducting the event detection and for reporting a corresponding result. Then, a determination is made regarding a modification of a configuration of the prediction functionality, e.g. a beam prediction using an AI / ML model; for example, it is decided whether, for example, the AI / ML model is to be activated or deactivated. In this context, also performance monitoring result can be used.

[0089] It is to be noted that events considered in examples of embodiments, i.e. input change events, are related to an case where an event is triggered without relying on the outcome / prediction of the model to respond and indicate the event to NW.

[0090] According to examples of embodiments, a UE-sided AI / ML model with performance monitoring procedure is considered wherein measures for determining the validity of beam prediction or performance monitoring validity are located on the NW side or the UE side.

[0091] In the following, it is assumed that the functionalities described to be located at the network side are implemented in a communication network control element or function, such as a gNB. However, examples of embodiments are not limited to a gNB as the NW side involved in the processing. Also, other network elements or functions can be used for executing processing parts described to be on the NW side, such as a core network element or function, a server, and the like. Furthermore, more than one network element or function can be used for executing processing parts described to be related to the NW side.

[0092] It is to be noted that the event to be considered can be of different types, such as a blockage detection, a blockage prediction, a detection of a rotation or motion of the UE, and the like, i.e. an event which causes an input distribution change (e.g. RSRP distribution) at the UE.

[0093] For detecting or determining the occurrence of such an event, according to examples of embodiments, different processes or means are applicable. For example, an internal or external sensor at the UE side or the network side (e.g. at the site of a gNB or the like) is employed that can detect or predict the event. As corresponding examples, LiDAR (Light detection and ranging) using on visual information, a camera, a motion sensor (e.g. ultrasonic or infrared), radar and accelerometers can be mentioned. That is, the UE or NW uses measurements and / or sensor results to detect an event, such as a blockage event, that may involve the activation / deactivation of an ML model.

[0094] Alternatively, or additionally, radio data, such as RSRP, is used on the UE side or the NW side so as to detect or predict events. That is, communication related parameters being measured on the UE side are usable for deciding, at the UE or the NW, whether an event is present.

[0095] In a first example of embodiments, for a UE-sided AI / ML model scenario, a case is considered where a UE-initiated procedure is applied. The UE supports beam prediction via a CSI report configuration (i.e. Set B), in which case the UE receives a configuration for monitoring the performance of the beam prediction. For example, the UE is configured for AI / ML model inference (with CSI-RS config_x) and performance monitoring (with CSI-RS config_y).

[0096] In the present example, it is assumed that the CSI reporting can be controlled (e.g. reconfigured / deactivated / switched). As an alternative, the UE supports a model level performance monitoring where, for example, the AI / ML model processing on the UE side can be controlled (e.g. activated / deactivated / switched).

[0097] In one example, the UE receives from the NW a configuration for the UE defining how to trigger an event. As an alternative, the corresponding configuration is preset on the UE side, e.g. by defining corresponding default settings.

[0098] As indicated above, there are different types of events, such as an event related to the blockage of signal reception. In this case, for example, it is possible to configure an event determination on the basis of radio data concerning parameters associated with the Set B measurements at the UE. For example, a corresponding parameter set considered for determining the event is configured in the configuration for event triggering, and comprises, for example, a threshold for L1-RSRP (referred to as R, for example) used for detecting the blockage, and time duration (referred to as T, for example) used for determining an applicable minimum blockage duration.

[0099] Alternatively, or additionally, in case a sensor measurement is used in connection with the event determination, e.g. in case the UE (or the NW element) supports sensing or predicting the blockage or the like on the basis of sensor information, a threshold to detect or predict a blockage gain (referred to as BG, for example) based on sensor information is part of the configuration (e.g. received from the NW). That is, the UE detects / predicts the event based on measurement and / or sensor data with respect to the configured thresholds R, T, BG., which are configured e.g. by the NW.

[0100] According to the present example of embodiments, when the UE predict / detects e.g. an L1- RSRP variation for received Set B measurements associated with the CSI report configuration and / or blockage gain derived based on sensor data, the UE triggers a suitable event notification to the NW.

[0101] That is, the UE determines, for example, which type of event has been detected or predicted, and includes a corresponding indication into the notification. For example, the type of the event is determined by using the radio data and / or sensor measurements which are interpreted so as to select which type of event occurred. Event types to be discriminated are, for example, the following:

[0102] ■ Eventl : a start of a blockage;

[0103] ■ Event2: an end of the blockage;

[0104] ■ Event3: a rotation of the UE; and

[0105] ■ Event4: a motion of the UE.

[0106] Depending on which event type has been found, a corresponding indication of the event by using a suitable event ID is added to the notification. According to other examples, it is also possible that no discrimination of different types of events is configured at the UE; in this case, only a simple event notification can be used.

[0107] In addition to the event indication as such, according to further example, the UE determines and adds further information. For example, event IDs for predicted events are available. Time information (e.g. time stamp information related to the beginning / end of the event, an indication of an occurrence of the event within a time window) can be added. Furthermore, RS resources identifiers associated with the one or more beams affected by the event can be included as further information. Corresponding resources are for instance reference signal resources, wherein usable communication paths are channel paths in the gNB-UE communication, which may correspond to different TX-RX beams configuration at each channel path. Another information being applicable is an indication of an expected drop of channel quality caused by the event.

[0108] For forwarding or transmitting the notification regarding the event to the NW, the UE uses, in one example of an embodiment, for example a MAC-CE message in uplink (UL) direction. For example, a MAC-CE bit field is mapped to event information, including for example information like type of event, time information, and corresponding RS resources identifiers. In another example, the UE transmit a first event notification in uplink control information (UCI), e.g. 1- bit / multibit scheduling request, to the NW. Further information can follow by a second UL message that includes event information details.

[0109] When the NW (e.g. a gNB) receives the event notification from the UE, the NW conducts a process so as to determine which measures are to be taken. For example, it is decided whether a reconfiguration or a switch (i.e. change from the current configuration to another, known configuration) of beam prediction CSI-reporting is to be done, or if the ML model on the UE side is to be activated / deactivated / switched etc. That is, measures to be taken can include various measures, depending e.g. on negotiations between the UE and the NW, which influences the beam prediction results received from the UE.

[0110] According to examples of embodiments, the determination of the measure to be taken is based on the received event notification alone, or the received event notification and one or more other information already received or received in a specified time, such as performance monitoring reports from the UE.

[0111] That is, in one example, it is assumed that the NW receives at least one performance monitoring report and at least one event notification from the UE. In this case, the NW determines the measure to be taken, e.g. switching / deactivation of the beam prediction CSI report or the AI / ML model, based on the event notification and one or more performance monitoring reports. According to a further example, the NW determines the activation of the beam prediction CSI report based on the event notification and one or more performance monitoring reports of an inactive UE-sided ML model.

[0112] In another example, the NW receives at least one event notification while any performance monitoring report is not received. For example, in case the UE has limited processing resources and / or deprioritizes the processing of the CSI report for monitoring, a corresponding performance monitoring report is missing. In this case, the determination of the measure to be taken is made under consideration of the information included in the event notification.

[0113] For example, according to an example of embodiments, the NW determines and configures an CSI-report activation / deactivation command based on the event received. If CSI-report for beam prediction is non-active, the NW determines and notifies an CSI-report activation command based on the event received. Alternatively, in case the CSI-report for beam prediction is active, the NW determines a CSI-report deactivation based on the event received. In another case, the NW receives at least one performance monitoring report but an event notification is not received. Here, the NW can rely, as indicated above, on performance monitoring feedback in a time when the UE does not indicate an event (such as in 0 - 3s and 7 - 10 s in Fig. 4).

[0114] In another example, the event is not be notified although event has occurred, for example if sensors are not calibrated accurately. In this situation, the NW makes a decision regarding measures of be taken, such as a configuration regarding activation / deactivation / switching, by relying on the PM reports.

[0115] According to further examples of embodiments, the NW waits for a time interval t before instructing the UE about the measure to be taken (e.g. CSI report activation / deactivation / switching). This allows to avoid, for example, a frequent CSI report activation / deactivation / switching.

[0116] It is to be noted that the above measures are described mainly to be related to a CSI report related measure. As already indicated above, however, measures to be taken can also be related to activation / deactivation / switching of a UE sided ML model, when the performance monitoring happens at the model level (where models are identified at the NW), when the performance monitoring is configured at model-level.

[0117] In a further example of embodiments, for a UE-sided AI / ML model scenario, a case is considered where a NW-initiated procedure is applied. Again, it is assumed that the UE supports beam prediction via a CSI report configuration (i.e. Set B), in which case the UE receives a configuration for monitoring the performance of the beam prediction. For example, the UE is configured for AI / ML model inference (with CSI-RS config_x) and performance monitoring (with CSI-RS config_y).

[0118] In the present example, it is assumed that the CSI reporting can be controlled (e.g. reconfigured / deactivated / switched). As an alternative, the UE supports a model level performance monitoring where, for example, the AI / ML model processing on the UE side can be controlled (e.g. activated / deactivated / switched).

[0119] As indicated above, in the present embodiment, the NW determines occurrence of an event, such as an event related to the blockage of signal reception, or another type of event (e.g. as indicated above). The event determination is done on the basis of radio data concerning parameters associated with the Set B measurements at the UE (e.g. R and T, as indicated above) or in case a sensor measurement is used in connection with the event determination, a threshold for detecting or predicting a blockage gain (i.e. BG). That is, the NW UE detects / predicts the event based on measurement and / or sensor data with respect to thresholds R, T, BG.

[0120] Similarly to the above embodiment, the NW conducts, when the event is detected or predicted, a process so as to determine which measures are to be taken. For example, it is decided whether a reconfiguration or a switch (i.e. change from the current configuration to another, known configuration) of beam prediction CSI-reporting is to be done, or if the ML model on the UE side is to be activated / deactivated / switched etc. That is, measures to be taken can include various measures, depending e.g. on negotiations between the UE and the NW, which influences the beam prediction results received from the UE.

[0121] According to examples of embodiments, the determination of the measure to be taken is based on the detected or predicted event alone, or the detected or predicted event and one or more other information already received or received in a specified time, such as performance monitoring reports from the UE.

[0122] That is, in one example, it is assumed that the NW detects receives at least one performance monitoring report and detects at least one event (e.g. from sensor measurements). In this case, the NWdetermines the measure to be taken, e.g. switching / deactivation of the beam prediction CSI report or the AI / ML model, based on the event notification and one or more performance monitoring reports.

[0123] For example, as an example scenario for explaining when switching / deactivation of the CSI report or AI / ML model is applicable, the following can be assumed.

[0124] The UE-side uses two models 1 and 2. It is assumed that model 1 performs well in nonblockage scenario but poorly in blocked scenarios, while model 2 is performs well in blockage scenarios compared model 1 but performs poorly in non-blocked scenarios compared to model 1.

[0125] In case model-level performance monitoring is performed, NW knows the existence of two models at UE and can indicate to the UE to switch the AI / ML model based on the event. On the other hand, when the NW does not know the existence of two models at the UE, the NW may indicate the UE to activate / switch CSI reporting based on the event.

[0126] According to a further example, the NW determines the activation of the beam prediction CSI report based on the event notification and one or more performance monitoring reports of an inactive UE-sided ML model.

[0127] In another example, the NW detects at least one event while any performance monitoring report is not received. In this case, the determination of the measure to be taken is made under consideration of the information obtained in connection with the event detection.

[0128] For example, according to an example of embodiments, the NW determines and configures an CSI-report activation / deactivation command based on the event detection. If CSI-report for beam prediction is non-active, the NW determines and notifies an CSI-report activation command based on the event detected. Alternatively, in case the CSI-report for beam prediction is active, the NW determines a CSI-report deactivation based on the event detected.

[0129] In another case, the NW receives at least one performance monitoring report, but an event is not detected. Here, the NW can rely, as indicated above, on performance monitoring feedback in a time when, for example, the sensor does not detect an event (such as in 0 - 3s and 7 - 10 s in Fig. 4).

[0130] In another example, the event is not be notified although event has occurred, for example if sensors are not calibrated accurately. In this situation, the NW makes a decision regarding measures of be taken, such as a configuration regarding activation / deactivation / switching, by relying on the PM reports.

[0131] According to further examples of embodiments, the NW waits for a time interval t before instructing the UE about the measure to be taken (e.g. CSI report activation / deactivation / switching). This allows to avoid, for example, a frequent CSI report activation / deactivation / switching.

[0132] It is to be noted that the above measures are described mainly to be related to a CSI report related measure. As already indicated above, however, measures to be taken can also be related to activation / deactivation / switching of a UE sided ML model, when the performance monitoring happens at the model level (where models are identified at the NW), when the performance monitoring is configured at model-level. In the following, examples of embodiments related to a processing according to the above principles are described. Specifically, in the following examples, an implementation is assumed in which a sensor, specifically a Lidar sensor, is used for detecting whether an event to be considered, such as a blockage event, occurs.

[0133] According to the presented examples of embodiments, Lidar technology is used to collect realtime 3D maps of the environment. The Lidar sensor transmits a set of rotating laser beams and measure the distance to the reflected surface based on the time-of-flight. The resulted 3D point cloud can be considered as a digital map of the physical environment. Known point cloud processing techniques / algorithms can be used to identify different objects in the environment such as human, tables, chairs etc. which can be a potential blockage for a considered UE in a wireless network. That is, in the following examples, it is assumed that the Lidar sensor, which is mounted, for example, at the UE or the gNB, captures the 3D point cloud of the environment and process or predicts the potential blockages in future. It is to be noted that also other sensor types, like radar sensor etc., can be employed in a similar manner, so that the Lidar sensor is to be seen only as a illustrative example.

[0134] According to the present examples of embodiments, the measurement results obtained from the Lidar sensor allows to determine whether a blockage or the like exists. Fig. 5 shows a diagram illustrating a resulting blockage gain determination. That is, Fig. 5 shows an example where it is determined, for example, whether the AI / ML model is deactivated or not on the basis of a predicted blockage gain (BG) using LiDAR data. When the blockage event is predicted (e.g. BG < 0.5), the an AI / ML model is deactivated, and if BG> 0.5, AI / ML model is kept active or re-activated. The dashed box shown in Fig. 5 shows a continuous duration where AI / ML model is deactivated (blockage event detected).

[0135] Fig. 6 shows an example of a procedure where, as a measure to be taken in case of an event triggering, a CSI-report reconfiguration / switching is done at the UE, wherein the event determination is based on a detection using a Lidar sensor on the UE side.

[0136] In S610, the NW configures the UE regarding CSI measurements with RS resources (SSB / CSI-RS resources).

[0137] In S620, the NW configures CSI-RS configuration which supports the beam prediction (config_x) and another CSI-RS configuration for performance monitoring purposes (config_y). In S630, the NW configures the UE on event notification. That is, the NW provides the UE with information regarding event detection or prediction, which includes for example information regarding event IDs and associated thresholds.

[0138] In S640, a CSI-RS transmission for AI / ML model inference (with CSI-RS config_x) or performance monitoring (with CSI-RS config_y) is executed.

[0139] In S650, the AI / ML model inference at the UE is executed for considering measurements associated with CSI-RS config_x as the input and predicting the strongest K resource IDs (best K beams) corresponding to CSI-RS config_Y. Then, in S660, the predicted best K beams are report to the NW.

[0140] In S665, the UE feedback performance monitoring based on the configured performance indicators and / or monitored RS resources.

[0141] In S670, the UE detects / predicts a potential blockage event on the basis of sensor measurements (Lidar measurements). According to the present example, the UE determines also associated information on the predicted blocker, such as an expected duration of the blockage, an expected RSRP degradation, potential beams to be impacted etc.

[0142] In S675, UE determines or selects, using the detected information, the event type to be notified to the NW based on the configuration provided by the NW.

[0143] In S680, the UE triggers the event notification with the relevant event ID and associated information.

[0144] In S690, the NW determines the measure to be taken in view of the event. That is, in the present example, the NW determines the CSI-report reconfiguration / switching based on information obtained from the UE, i.e. the received performance monitoring report and the received event notification.

[0145] In S695, in case the NW decided to reconfigure or switch the CSI-report, the NW notifies or instructs the UE to reconfigure or switch CSI-report.

[0146] Fig. 7 shows an example of a procedure where, as a measure to be taken in case of an event triggering, a AI / ML model activation / deactivation is done at the UE, wherein the event determination is based on a detection using a Lidar sensor on the UE side, wherein modellevel performance monitoring is executed.

[0147] In S710, the NW configures the UE regarding CSI measurements with RS resources (SSB / CSI-RS resources).

[0148] In S720, the NW configures CSI-RS configuration which supports the beam prediction (config_x) and another CSI-RS configuration for model level performance monitoring purposes (config_y).

[0149] In S730, the NW configures the UE on event notification. That is, the NW provides the UE with information regarding event detection or prediction, which includes for example information regarding event IDs and associated thresholds.

[0150] In 7640, a CSI-RS transmission for AI / ML model inference (with CSI-RS config_x) or performance monitoring (with CSI-RS config_y) is executed.

[0151] In S750, the AI / ML model inference at the UE is executed for considering measurements associated with CSI-RS config_x as the input and predicting the strongest K resource IDs (best K beams) corresponding to CSI-RS config_Y. Then, in S760, the predicted best K beams are report to the NW.

[0152] In S765, the UE reports performance of the AI / ML model(s) (it is to be noted that more than one model can be used at the same time) based on the configured performance indicators.

[0153] In S770, the UE detects / predicts a potential blockage event on the basis of sensor measurements (Lidar measurements). According to the present example, the UE determines also associated information on the predicted blocker, such as an expected duration of the blockage, an expected RSRP degradation, potential beams to be impacted etc.

[0154] In S775, UE determines or selects, using the detected information, the event type to be notified to the NW based on the configuration provided by the NW.

[0155] In S780, the UE triggers the event notification with the relevant event ID and associated information. In S790, the NW determines the measure to be taken in view of the event. That is, in the present example, the NW determines AI / ML model deactivation / switching based on the information obtained from the UE, i.e. the received model-level performance monitoring report and the received event notification.

[0156] In S795, in case the NW decided to deactivate or switch the AI / ML model, the NW deactivates or switches the AI / ML model.

[0157] Fig. 8 shows an example of a procedure where, as a measure to be taken in case of an event triggering, a CSI-report reconfiguration / switching is done at the UE, wherein the event determination is based on a detection using a Lidar sensor on the NW side.

[0158] In S810, the NW configures the UE regarding CSI measurements with RS resources (SSB / CSI-RS resources).

[0159] In S820, the NW configures CSI-RS configuration which supports the beam prediction (config_x) and another CSI-RS configuration for performance monitoring purposes (config_y).

[0160] In S830, a CSI-RS transmission for AI / ML model inference (with CSI-RS config_x) or performance monitoring (with CSI-RS config_y) is executed.

[0161] In S840, the AI / ML model inference at the UE is executed for considering measurements associated with CSI-RS config_x as the input and predicting the strongest K resource IDs (best K beams) corresponding to CSI-RS config_Y. Then, in S850, the predicted best K beams are report to the NW.

[0162] In S855, the UE feedback performance monitoring based on the configured performance indicators and / or monitored RS resources.

[0163] In S860, the NW detects / predicts a potential blockage event on the basis of sensor measurements (Lidar measurements). According to the present example, the NW determines also associated information on the predicted blocker, such as an expected duration of the blockage, an expected RSRP degradation, potential beams to be impacted etc.

[0164] In S870, the NW determines the measure to be taken in view of the event. That is, in the present example, the NW determines the CSI-report reconfiguration / switching based on information obtained from the UE, i.e. the received performance monitoring report, and the event detected.

[0165] In S880, in case the NW decided to reconfigure or switch the CSI-report, the NW notifies or instructs the UE to reconfigure or switch CSI-report.

[0166] Fig. 9 shows an example of a procedure where, as a measure to be taken in case of an event triggering, a AI / ML model activation / deactivation is done at the UE, wherein the event determination is based on a detection using a Lidar sensor on the NW side, wherein modellevel performance monitoring is executed.

[0167] In S910, the NW configures the UE regarding CSI measurements with RS resources (SSB / CSI-RS resources).

[0168] In S920, the NW configures CSI-RS configuration which supports the beam prediction (config_x) and another CSI-RS configuration for model level performance monitoring purposes (config_y).

[0169] In S930, a CSI-RS transmission for AI / ML model inference (with CSI-RS config_x) or performance monitoring (with CSI-RS config_y) is executed.

[0170] In S940, the AI / ML model inference at the UE is executed for considering measurements associated with CSI-RS config_x as the input and predicting the strongest K resource IDs (best K beams) corresponding to CSI-RS config_Y. Then, in S950, the predicted best K beams are report to the NW.

[0171] In S955, the UE reports performance of the AI / ML model(s) (it is to be noted that more than one model can be used at the same time) based on the configured performance indicators.

[0172] In S960, the NW detects / predicts a potential blockage event on the basis of sensor measurements (Lidar measurements). According to the present example, the NW determines also associated information on the predicted blocker, such as an expected duration of the blockage, an expected RSRP degradation, potential beams to be impacted etc.

[0173] In S970, the NW determines the measure to be taken in view of the event. That is, in the present example, the NW determines AI / ML model deactivation / switching based on the information obtained from the UE, i.e. the received model-level performance monitoring report and the received event notification.

[0174] In S980, in case the NW decided to deactivate or switch the AI / ML model, the NW deactivates or switches the AI / ML model.

[0175] Fig. 10 shows a flow chart of a processing for controlling a communication state management process in a communication network as shown in Fig. 1. That is, the flow chart shown in Fig. 10 illustrates a processing related to a communication state management process which is conducted on the network side, such as by a gNB.

[0176] In S1010, the NW determines an input change event. The input change event is related to a change of power in a received signal at a UE capable of using a prediction functionality for predicting a communication state.

[0177] As indicated above, according to examples of embodiments, the input change event is related to a situation where the event is triggered without relying on the outcome / prediction of the model to respond and indicate the event to the NW.

[0178] According to examples of embodiments, predicting a communication state means, for example, a beam prediction, a prediction of a channel state information in a communication with the UE, a prediction of a received signal strength in a communication with the UE, or the like.

[0179] Furthermore, according to examples of embodiments, the input change event is related to one of the following: a start of a blockage of a communication path to the UE, a stop of a blockage of a communication path to the UE, a movement of the UE, or a rotation of the UE.

[0180] For determining the input change event, according to examples of embodiments, the NW detects or predicts the occurrence of the input change event. In this case, according to examples of embodiments, the occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data obtained from at least one sensor are used, wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value. Alternatively, or additionally, radio data obtained for the UE (i.e. measured at the NW, or measured and reported by the UE) are used, wherein the input change event is detected or predicted by comparing values measured in a communication with the UE with a preset communication threshold value and a time threshold value.

[0181] According to examples of embodiments, the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following: a Lidar sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor. Furthermore, according to examples of embodiments, the radio data comprise a received signal power value on the user equipment side and a time information relating to a change in the received signal power.

[0182] On the other hand, for determining the input change event, according to further examples of embodiments, the NW receives an event notification from the UE. In this case, according to examples of embodiments, the event notification received from the UE comprises information indicating that the UE has detected or predicted the occurrence of the input change event.

[0183] According to examples of embodiments, the event notification received from the UE comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event.

[0184] In S1020, the NW determines whether a modification of a configuration of the prediction functionality is required. This determination is made based at least in part on the determination of the input change event (that is, for example, based on information or data obtained by the NW, information or data received in connection with the event notification from the UE, information or data obtained by other means, such as a performance monitoring report, and the like).

[0185] According to examples of embodiments, the determination as to whether the modification of the configuration of the prediction functionality is required is based on at least one of the following: information about a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; information about communication resources associated with one or more communication paths affected by the event (for example, communication resources are reference signal resources and communication paths are channel paths in the gNB-UE communication, which may correspond to different TX-RX beams configuration at each channel path); information about an expected drop of communication quality caused by the event, or at least one of performance monitoring report indicating a communication state prediction performance on the user equipment side.

[0186] Moreover, according to examples of embodiments, the modification of the configuration of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting; an activation or deactivation of a ML based communication state prediction model (e.g. of a ML model for beam prediction), or a switching to a different ML based communication state prediction model.

[0187] In S1030, in response to a determination that the modification is required, the NW instructs the UE to modify the configuration of the prediction functionality.

[0188] According to further examples of embodiments, the NW waits, after it is determined that a modification of the configuration of the prediction functionality is required, a preset time interval and instructs the user equipment to modify the configuration of the prediction functionality after the waiting time interval has expired.

[0189] Furthermore, according to examples of embodiments, the NW configures the UE for triggering the input change event, and also provides information for detecting or predicting the input change event and for preparing an event indication (the event notification) at the UE side.

[0190] Furthermore, according to examples of embodiments, the NW receives, from the UE, channel state information prediction reports and performance monitoring reports. The NW can use at least one of the received channel state information prediction reports and performance monitoring reports for determining whether a modification of the configuration of the prediction functionality is required.

[0191] Fig. 11 shows a flow chart of a processing for controlling a communication state management process in a communication network as shown in Fig. 1. That is, the flow chart shown in Fig. 11 illustrates a processing related to a communication state management process which is conducted on the UE side, such as by an UE. According to examples of embodiments, the processing is executed in a UE or in a part attachable to the UE. In S1110, the UE detects or predicts an input change event. The input change event is related to a change of power in a received signal at a UE capable of using a prediction functionality for predicting a communication state. The occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data received from at least one sensor, or radio data measured at the user equipment.

[0192] According to examples of embodiments, predicting a communication state means, for example, a beam prediction, a prediction of a channel state information of a communication path with the UE, a prediction of a received signal strength in a communication with the UE, or the like.

[0193] Furthermore, according to examples of embodiments, the input change event is related to one of the following: a start of a blockage of a communication path to the UE, a stop of a blockage of a communication path to the UE, a movement of the UE, or a rotation of the UE.

[0194] According to examples of embodiments, the UE detects or predicts the occurrence of the input change event on the basis of the sensor data by comparing a preset sensor threshold value with a sensor output value. Alternatively, or additionally, the input change event is detected or predicted by comparing values measured in a communication with the UE with a preset communication threshold value and a time threshold value.

[0195] According to examples of embodiments, the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following: a Lidar sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor. Furthermore, according to examples of embodiments, the radio data comprise a received signal power value at the UE and a time information relating to a change in the received signal power.

[0196] In S1120, in response to that the input change event is detected or predicted, the UE prepares an event notification, which is then sent in S1130 to a communication network.

[0197] According to examples of embodiments, the event notification sent to the communication network comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event. Furthermore, according to examples of embodiments, the UE receives configuration information from the communication network for triggering the input change event, and information for detecting or predicting the input change event (such as thresholds and the like) and for preparing an event notification (such as information regarding types of events).

[0198] According to examples of embodiments, the UE transmits, to the communication network, also channel state information prediction reports and performance monitoring reports related to the prediction functionality.

[0199] According to further examples of embodiments, the UE receives an instruction for modifying the configuration of the prediction functionality. In this case, the UE executes the modification of the configuration of the prediction functionality. The modification of the configuration of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting; an activation or deactivation of a ML based communication state prediction model, or a switching to a different ML based communication state prediction model.

[0200] Fig. 12 shows a diagram of a communication network control element 20, such as a gNB, which conducts a processing related to a communication state management processing, such as a beam management processing, according to some examples of the embodiments, as described in connection with Figs. 5 to 9. It is to be noted that the communication network control element 20, which is configured to be a gNB, may include further elements or functions besides those described herein below. Furthermore, even though reference is made to communication network control element or function, the element or function may be also another device or function having a similar task, such as a chipset, a chip, a module, an application etc., which can also be part of another element or attached as a separate element to another element, or the like. It should be understood that each block and any combination thereof may be implemented by various means or their combinations, such as hardware, software, firmware, one or more processors and / or circuitry.

[0201] The communication network control element 20 shown in Fig. 12 may include a processing circuitry, a processing function, a control unit or a processor 201 , such as a CPU or the like, which is suitable for executing instructions given by programs or the like related to the control procedure. The processor 201 may include one or more processing portions or functions dedicated to specific processing as described below, or the processing may be run in a single processor or processing function. Portions for executing such specific processing may be also provided as discrete elements or within one or more further processors, processing functions or processing portions, such as in one physical processor like a CPU, for example. Reference signs 202 and 203 denote input / output (I / O) units or functions (interfaces) connected to the processor or processing function 201. The I / O units 202 may be used for communicating with a user equipment. The I / O units 203 may be used for communicating with other network elements, such as a core network element or function of a communication network. The I / O units 202 and 203 may be combined units including communication equipment towards several entities, or may include a distributed structure with a plurality of different interfaces for different entities. Reference sign 204 denotes a memory usable, for example, for storing data and programs to be executed by the processor or processing function 201 and / or as a working storage of the processor or processing function 201 . It is to be noted that the memory 204 may be implemented by using one or more memory portions of the same or different type of memory.

[0202] The processor or processing function 201 is configured to execute processing related to the above described control procedure. In particular, the processor or processing circuitry or function 201 includes one or more of the following sub-portions. Sub-portion 2011 is a processing portion which is usable as a portion for determining an input change event. The portion 2011 may be configured to perform processing according to S1010 of Fig. 10. Furthermore, the processor or processing circuitry or function 201 may include a sub-portion 2012 usable as a portion for determining whether a modification of a configuration of a prediction functionality is required. The portion 2012 may be configured to perform a processing according to S1020 of Fig. 10. In addition, the processor or processing circuitry or function 201 may include a sub-portion 2013 usable as a portion for instructing a modification of the configuration of the prediction functionality. The portion 2013 may be configured to perform a processing according to S1030 of Fig. 10.

[0203] Fig. 13 shows a diagram of a communication element 10, such as a UE, which conducts a processing related to a communication state management processing, such as a beam management processing, according to some examples of the embodiments, as described in connection with Figs. 5 to 9. It is to be noted that the communication element 10, which is configured to be a UE, may include further elements or functions besides those described herein below. Furthermore, even though reference is made to network element or function, the element or function may be also another device or function having a similar task, such as a chipset, a chip, a module, an application etc., which can also be part of another element or attached as a separate element to another element, or the like. It should be understood that each block and any combination thereof may be implemented by various means or their combinations, such as hardware, software, firmware, one or more processors and / or circuitry.

[0204] The communication element 10 shown in Fig. 13 may include a processing circuitry, a processing function, a control unit or a processor 101 , such as a CPU or the like, which is suitable for executing instructions given by programs or the like related to the control procedure. The processor 101 may include one or more processing portions or functions dedicated to specific processing as described below, or the processing may be run in a single processor or processing function. Portions for executing such specific processing may be also provided as discrete elements or within one or more further processors, processing functions or processing portions, such as in one physical processor like a CPU, for example. Reference sign 102 denote input / output (I / O) units or functions (interfaces) connected to the processor or processing function 101. The I / O unit 102 may be used for communicating with a network, such as a gNB. The I / O units 102 may be combined units including communication equipment towards several entities, or may include a distributed structure with a plurality of different interfaces for different entities. Reference sign 104 denotes a memory usable, for example, for storing data and programs to be executed by the processor or processing function 101 and / or as a working storage of the processor or processing function 101. It is to be noted that the memory 104 may be implemented by using one or more memory portions of the same or different type of memory.

[0205] The processor or processing function 101 is configured to execute processing related to the above described control procedure. In particular, the processor or processing circuitry or function 101 includes one or more of the following sub-portions. Sub-portion 1011 is a processing portion which is usable as a portion for detecting or predicting an input change event. The portion 1011 may be configured to perform processing according to S1110 of Fig. 11. Furthermore, the processor or processing circuitry or function 101 may include a subportion 1012 usable as a portion for preparing an event notification. The portion 1012 may be configured to perform a processing according to S1120 of Fig. 11. In addition, the processor or processing circuitry or function 101 may include a sub-portion 1013 usable as a portion for sending the event notification. The portion 1013 may be configured to perform a processing according to S1130 of Fig. 11.

[0206] According to a further example of embodiments, in an aspect 1 , there is provided, for example, an apparatus comprising means for determining an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, means for determining whether a modification of a configuration of the prediction functionality is required, based at least in part on the determination of the input change event, and means for instructing, in response to a determination that the modification is required, the user equipment to modify the prediction functionality.

[0207] Furthermore, according to some other examples of embodiments, the above defined apparatus may further comprise means for conducting at least one of the processing defined in the above described methods, for example a method according to that described in connection with Fig. 10.

[0208] Specifically, in an aspect 2, the apparatus according to aspect 1 is provided, wherein to determine the input change event comprises one of the following: to detect or predict occurrence of the input change event, or to receive an event notification from the user equipment, wherein to detect or predict the occurrence of the input change event is based on at least one of the following: sensor data obtained from at least one sensor, wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value, or radio data obtained for the user equipment, wherein the input change event is detected or predicted by comparing values measured in a communication with the user equipment with a preset communication threshold value and a time threshold value; and wherein the event notification received from the user equipment comprises information indicating that the user equipment has detected or predicted the occurrence of the input change event.

[0209] In an aspect 3, the apparatus according to aspect 2 is provided, wherein the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following: a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power.

[0210] In an aspect 4, the apparatus according to aspect 2 is provided, wherein the event notification received from the user equipment comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event. In an aspect 5, the apparatus according to any of aspects 1 to 4 is provided, wherein the input change event comprises one of the following: a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment; and wherein the prediction functionality is for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment.

[0211] In an aspect 6, the apparatus according to any of aspects 1 to 5 is provided, further comprising means for configuring the user equipment for triggering the input change event, and means for providing information for detecting or predicting the input change event and for preparing the event notification at the user equipment side.

[0212] In an aspect 7, the apparatus according to aby of aspects 1 to 6 is provided, further comprising means for receiving, from the user equipment, channel state information prediction reports and performance monitoring reports, and means for using at least one of the received channel state information prediction reports and performance monitoring reports for determining whether a modification of the configuration of the prediction functionality is required.

[0213] In an aspect 8, the apparatus according to any of aspects 1 to 7 is provided, wherein: determining whether the modification of the configuration of the prediction functionality is required, is based on at least one of the following: information about a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; information about communication resources associated with one or more communication paths affected by the event; information about an expected drop of communication quality caused by the event, or at least one of performance monitoring report indicating a communication state prediction performance on the user equipment side.

[0214] In an aspect 9, the apparatus according to any of aspects 1 to 8 is provided, wherein the modification of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting, an activation or deactivation of a machine learning based communication state prediction model, or a switching to a different machine learning based communication state prediction model.

[0215] In an aspect 10, the apparatus according to any of aspects 1 to 9 is provided, further comprising means for waiting, after it is determined that a modification of the configuration of the prediction functionality is required, a preset time interval and means for instructing the user equipment to modify the configuration of the prediction functionality after the waiting time interval has expired.

[0216] According to a further example of embodiments, there is provided, for example, a non- transitory computer readable medium comprising program instructions for causing an apparatus to perform a processing comprising determining an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, determining whether a modification of a configuration of the prediction functionality is required, based at least in part on the determination of the input change event, and in response to a determination that the modification is required, instructing the user equipment to modify the prediction functionality.

[0217] According to a further example of embodiments, in an aspect 11 , there is provided, for example, an apparatus comprising means for detecting or predicting an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, wherein the occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data received from at least one sensor, or radio data measured at the user equipment; means for preparing, in response to that the input change event is detected or predicted, an event notification, and means for sending, to a communication network, the prepared event notification.

[0218] Furthermore, according to some other examples of embodiments, the above defined apparatus may further comprise means for conducting at least one of the processing defined in the above described methods, for example a method according to that described in connection with Fig. 11.

[0219] Specifically, in an aspect 12, the apparatus according to aspect 11 is provided, wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value, or the input change event is detected or predicted by comparing values measured for a communication with the communication network with a preset communication threshold value and a time threshold value.

[0220] In an aspect 13, the apparatus according to aspect 11 or 12 is provided, wherein the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following: a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power.

[0221] In an aspect 14, the apparatus according to any of aspects 11 to 13 is provided, wherein the event notification sent to the communication network comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event.

[0222] In an aspect 15, the apparatus according to any of aspects 11 to 14 is provided, wherein the input change event comprises one of the following: a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment, and wherein the prediction functionality is for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment.

[0223] In an aspect 16, the apparatus according to any of aspects 11 to 15 is provided, further comprising means for receiving configuration information from the communication network for triggering the input change event, and means for receiving information for detecting or predicting the input change event and for preparing the event notification .

[0224] In an aspect 17, the apparatus according to any of aspects 11 to 16 is provided, further comprising means for transmitting, to the communication network, channel state information prediction reports and performance monitoring reports related to the prediction functionality. In an aspect 18, the apparatus according to any of aspects 11 to 17 is provided, further comprising means for receiving an instruction for modifying the prediction functionality, and means for executing the modification of the prediction functionality, wherein the modification of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting; an activation or deactivation of a machine learning based communication state prediction model, or a switching to a different machine learning based communication state prediction model.

[0225] In an aspect 19, the apparatus according to any of aspects 11 to 18 is provided, wherein the apparatus is comprised in the user equipment or attachable to the user equipment.

[0226] According to a further example of embodiments, there is provided, for example, a non- transitory computer readable medium comprising program instructions for causing an apparatus to perform a processing comprising detecting or predicting an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, wherein the occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data received from at least one sensor, or radio data measured at the user equipment; in response to that the input change event is detected or predicted, preparing an event notification, and sending, to a communication network, the prepared event notification.

[0227] By means of the procedures described above, it is possible to accelerate a reconfiguration of a UE with regard to a provision of prediction functionality results. Specifically, it is possible, when the UE sends an event notification, that the UE is reconfigured more quickly than in a legacy scenario, where the UE has to wait that the NW recognize the ML model performance drop based on the received monitoring report.

[0228] For example, considering an example of performance monitoring in legacy scenario,

[0229] For example, 6JKP may be determined over a “certain number of PM instances” to capture a normalized behavior of PM events of the model. If the total number of instances considered to calculateesssp is N, the PM report is received at the NW after N PM instances. This delays to capture effect of the input change event. Additionally, unlike beam prediction report, PM reporting does not have strict latency requirements. These cases allow the ML model to make poor predictions over a long duration. On the other hand, according to examples of embodiments of the present invention, the implementation of input change event indication allows a UE to send the notification even before the ML model starts to perform poorly. That is, the transmission of the event notification can be used as a trigger for a fast reconfiguration, i.e. receiving a reconfiguration setting earlier than in a legacy scenario.

[0230] Furthermore, it is possible to have a low signaling overhead for modifying a configuration of a prediction functionality. For example, when considering AI / ML model activation / deactivation, it is possible to avoid a high frequency regarding switching between activation and deactivation of the AI / ML model, so that the signaling load in the network can be reduced.

[0231] In addition, by achieving a reliable model activation and deactivation, it is possible to obtain a higher user throughput for blocked UEs.

[0232] According to an aspect 21 , there is provided a method comprising determining an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, determining whether a modification of a configuration of the prediction functionality is required, based at least in part on the determination of the input change event, and in response to a determination that the modification is required, instructing the user equipment to modify the configuration of the prediction functionality.

[0233] According to an aspect 22, the method according to aspect 21 is provided, wherein to determine the input change event comprises one of the following: to detect or predict occurrence of the input change event, or to receive an event notification from the user equipment, wherein to detect or predict the occurrence of the input change event is based on at least one of the following: sensor data obtained from at least one sensor, wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value, or radio data obtained for the user equipment, wherein the input change event is detected or predicted by comparing values measured in a communication with the user equipment with a preset communication threshold value and a time threshold value; and wherein the event notification received from the user equipment comprises information indicating that the user equipment has detected or predicted the occurrence of the input change event. According to an aspect 23, the method according to aspect 22 is provided, wherein the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following: a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power.

[0234] According to an aspect 24, the method according to aspect 22 is provided, wherein the event notification received from the user equipment comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event.

[0235] According to an aspect 25, the method according to any of aspects 21 to 24 is provided, wherein the input change event comprises one of the following: a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment; and wherein the prediction functionality is for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment.

[0236] According to an aspect 26, the method according to any of aspects 21 to 25 is provided, further comprising configuring the user equipment for triggering the input change event, and providing information for detecting or predicting the input change event and for preparing the event notification at the user equipment side .

[0237] According to an aspect 27, the method according to any of aspects 21 to 26 is provided, further comprising receiving, from the user equipment, channel state information prediction reports and performance monitoring reports, and using at least one of the received channel state information prediction reports and performance monitoring reports for determining whether a modification of the configuration of the prediction functionality is required.

[0238] According to an aspect 28, the method according to any of aspects 21 to 28 is provided, wherein determining whether the modification of the configuration of the prediction functionality is required, is based on at least one of the following: information about a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; information about communication resources associated with one or more communication paths affected by the event; information about an expected drop of communication quality caused by the event, or at least one of performance monitoring report indicating a communication state prediction performance on the user equipment side.

[0239] According to an aspect 29, the method according to any of aspects 21 to 28 is provided, wherein the modification of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting, an activation or deactivation of a machine learning based communication state prediction model, or a switching to a different machine learning based communication state prediction model.

[0240] According to an aspect 30, the method according to any of aspects 21 to 29 is provided, further comprising waiting, after it is determined that a modification of the configuration of the prediction functionality is required, a preset time interval and instructing the user equipment to modify the configuration of the prediction functionality after the waiting time interval has expired.

[0241] According to an aspect 31 , there is provided a method comprising detecting or predicting an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, wherein the occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data received from at least one sensor, or radio data measured at the user equipment; in response to that the input change event is detected or predicted, preparing an event notification, and sending, to a communication network, the prepared event notification

[0242] According to an aspect 32, the method according to aspect 31 is provided, wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value, or the input change event is detected or predicted by comparing values measured for a communication with the communication network with a preset communication threshold value and a time threshold value. According to an aspect 33, the method according to aspect 31 or 32 is provided, wherein the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following: a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power.

[0243] According to an aspect 34, the method according to any of aspects 31 to 33 is provided, wherein the event notification sent to the communication network comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event.

[0244] According to an aspect 35, the method according to any of aspects 31 to 34 is provided, wherein the input change event comprises one of the following: a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment, and wherein the prediction functionality is for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment.

[0245] According to an aspect 36, the method according to any of aspects 31 to 35 is provided, further comprising receiving configuration information from the communication network for triggering the input change event, and receiving information for detecting or predicting the input change event and for preparing the event notification .

[0246] According to an aspect 37, the method according to any of aspects 31 to 36 is provided, further comprising transmitting, to the communication network, channel state information prediction reports and performance monitoring reports related to the prediction functionality.

[0247] According to an aspect 38, the method according to any of aspects 31 to 37 is provided, further comprising receiving an instruction for modifying the prediction functionality, and executing the modification of the prediction functionality, wherein the modification of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting, an activation or deactivation of a machine learning based communication state prediction model, or a switching to a different machine learning based communication state prediction model.

[0248] According to an aspect 39, the method according to any of aspects 31 to 38 is provided, wherein the method is conducted in the user equipment or in a part attachable to the user equipment.

[0249] It should be appreciated that

[0250] - an communication technology via which data is transferred to and from an entity in the system may be any suitable present or future technology, such as WLAN (Wireless Local Access Network), WiMAX (Worldwide Interoperability for Microwave Access), LTE, LTE-A, 5G, Bluetooth, Infrared, and the like may be used; additionally, embodiments may also apply wired technologies, e.g. IP based access technologies like cable networks or fixed lines;

[0251] - embodiments suitable to be implemented as software code or portions of it and being run using a processor or processing function are software code independent and can be specified using any known or future developed programming language, such as a high-level programming language, such as objective-C, C, C++, C#, Java, Python, Javascript, other scripting languages etc., or a low-level programming language, such as a machine language, or an assembler;

[0252] - implementation of embodiments is hardware independent and may be implemented using any known or future developed hardware technology or any hybrids of these, such as a microprocessor or CPU (Central Processing Unit), MOS (Metal Oxide Semiconductor), CMOS (Complementary MOS), BiMOS (Bipolar MOS), BiCMOS (Bipolar CMOS), ECL (Emitter Coupled Logic), and / or TTL (Transistor-Transistor Logic);

[0253] - embodiments may be implemented as individual devices, apparatuses, units, means or functions, or in a distributed fashion, for example, one or more processors or processing functions may be used or shared in the processing, or one or more processing sections or processing portions may be used and shared in the processing, wherein one physical processor or more than one physical processor may be used for implementing one or more processing portions dedicated to specific processing as described;

[0254] - an apparatus may be implemented by a semiconductor chip, a chipset, or a (hardware) module including such chip or chipset;

[0255] - embodiments may also be implemented as any combination of hardware and software, such as ASIC (Application Specific IC (Integrated Circuit)) components, FPGA (Field-programmable Gate Arrays) or CPLD (Complex Programmable Logic Device) components or DSP (Digital Signal Processor) components;

[0256] - embodiments may also be implemented as computer program products, including a computer usable medium having a computer readable program code embodied therein, the computer readable program code adapted to execute a process as described in embodiments, wherein the computer usable medium may be a non-transitory medium.

[0257] Although the present disclosure has been described herein before with reference to particular embodiments thereof, the present disclosure is not limited thereto and various modifications can be made thereto.

Claims

CLAIMS1. An apparatus comprising at least one processor, and at least one memory for storing instructions that, when executed by the at least one processor, cause the apparatus at least to: determine an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, determine whether a modification of a configuration of the prediction functionality is required, based at least in part on the determination of the input change event, and in response to a determination that the modification is required, instruct the user equipment to modify the configuration of the prediction functionality.

2. The apparatus according to claim 1 , wherein to determine the input change event comprises one of the following: to detect or predict occurrence of the input change event, or to receive an event notification from the user equipment, wherein to detect or predict the occurrence of the input change event is based on at least one of the following: sensor data obtained from at least one sensor, wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value, or radio data obtained for the user equipment, wherein the input change event is detected or predicted by comparing values measured in a communication with the user equipment with a preset communication threshold value and a time threshold value; and wherein the event notification received from the user equipment comprises information indicating that the user equipment has detected or predicted the occurrence of the input change event.

3. The apparatus according to claim 2, wherein the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following:a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power.

4. The apparatus according to claim 2, wherein the event notification received from the user equipment comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event.

5. The apparatus according to any of claims 1 to 4, wherein the input change event comprises one of the following: a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment; and wherein the prediction functionality is for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment.

6. The apparatus according to any of claims 1 to 5, wherein the at least one processor and the at least one memory for storing instructions, when executed by the at least one processor, further cause the apparatus at least to: configure the user equipment for triggering the input change event, and provide information for detecting or predicting the input change event and for preparing the event notification at the user equipment side.

7. The apparatus according to any of claims 1 to 6, wherein the at least one processor and the at least one memory for storing instructions, when executed by the at least one processor, further cause the apparatus at least to:receive, from the user equipment, channel state information prediction reports and performance monitoring reports, and use at least one of the received channel state information prediction reports and performance monitoring reports for determining whether a modification of the configuration of the prediction functionality is required.

8. The apparatus according to any of claims 1 to 7, wherein: determining whether the modification of the configuration of the prediction functionality is required, is based on at least one of the following: information about a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; information about communication resources associated with one or more communication paths affected by the event; information about an expected drop of communication quality caused by the event, or at least one of performance monitoring report indicating a communication state prediction performance on the user equipment side.

9. The apparatus according to any of claims 1 to 8, wherein the modification of the configuration of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting; an activation or deactivation of a machine learning based communication state prediction model, or a switching to a different machine learning based communication state prediction model.

10. The apparatus according to any of claims 1 to 9, wherein the at least one processor and the at least one memory for storing instructions, when executed by the at least one processor, further cause the apparatus at least to: wait, after it is determined that a modification of the configuration of the prediction functionality is required, a preset time interval and instruct the user equipment to modify the configuration of the prediction functionality after the waiting time interval has expired.

11. An apparatus comprising at least one processor, and at least one memory for storing instructions that, when executed by the at least one processor, cause the apparatus at least to:detect or predict an input change event related to a change of power in a received signal at a user equipment capable of using a prediction functionality for predicting a communication state, wherein the occurrence of the input change event is detected or predicted on the basis of at least one of the following: sensor data received from at least one sensor, or radio data measured at the user equipment; in response to that the input change event is detected or predicted, prepare an event notification, and send, to a communication network, the prepared event notification.

12. The apparatus according to claim 11 , wherein the input change event is detected or predicted by comparing a preset sensor threshold value with a sensor output value, or the input change event is detected or predicted by comparing values measured for a communication with the communication network with a preset communication threshold value and a time threshold value.

13. The apparatus according to claim 11 or 12, wherein the at least one sensor from which the sensor data for detecting or predicting the occurrence of the input change event are obtained comprises at least one of the following: a LiDAR sensor, a radar sensor, a camera sensor, an acceleration sensor, or a motion sensor; and the radio data comprise a received signal power value at the user equipment and a time information relating to a change in the received signal power.

14. The apparatus according to any of claims 11 to 13, wherein the event notification sent to the communication network comprises at least one of the following: an identification of a type of the input change event; time information indicating at least one of a beginning of the event, an ending of the event, or an indication of an occurrence within a time window; an indication of communication resources associated with one or more communication paths affected by the event; or an indication of an expected drop of communication quality caused by the event.

15. The apparatus according to any of claims 11 to 14, wherein the input change event comprises one of the following: a start of a blockage of a communication path to the user equipment, a stop of a blockage of a communication path to the user equipment, a movement of the user equipment, or a rotation of the user equipment, and wherein the prediction functionality is for predicting at least one of the following: at least one beam used in a communication with the user equipment, a channel state information of a communication path with the user equipment, or a received signal strength in a communication with the user equipment.

16. The apparatus according to any of claims 11 to 15, wherein the at least one processor and the at least one memory for storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive configuration information from the communication network for triggering the input change event, and receive information for detecting or predicting the input change event and for preparing the event notification.

17. The apparatus according to any of claims 11 to 16, wherein the at least one processor and the at least one memory for storing instructions, when executed by the at least one processor, further cause the apparatus at least to: transmit, to the communication network, channel state information prediction reports and performance monitoring reports related to the prediction functionality.

18. The apparatus according to any of claims 11 to 17, wherein the at least one processor and the at least one memory for storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive an instruction for modifying a configuration of the prediction functionality, and execute the modification of the configuration of the prediction functionality, wherein the modification of the configuration of the prediction functionality comprises one of the following: a reconfiguration of a channel state information prediction reporting, an activation or deactivation of a channel state information prediction reporting, a switching to a different channel state information prediction reporting; an activation or deactivation of a machine learning based communication state prediction model, ora switching to a different machine learning based communication state prediction model.

19. The apparatus according to any of claims 11 to 18, wherein the apparatus is comprised in the user equipment or attachable to the user equipment.

Images