Prediction accuracy for wireless network mobility

Abstract

Multi-step processes may be used to obtain different predictions. For example, a multi-step process may comprise at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions. A wireless network may transmit a configuration for the multi-step process, and one or more apparatuses may receive the configuration. The apparatus may obtain measurement results of reference signals received by the apparatus and provide the results as input to the first process; and adjust, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

Description

[0001] DESCRIPTION

[0002] TITLE

[0003] PREDICTION ACCURACY FOR WIRELESS NETWORK MOBILITY

[0004] TECHNICAL FIELD

[0005] Various example embodiments relate to communication systems.

[0006] BACKGROUND

[0007] Communication systems are under constant development. The 5G, 5G-Advanced, and beyond future wireless networks, or network generations, aim to support a large variety of services, use cases and industrial verticals. Some of these may use predictions, for example mobility predictions, generated by a multi- step process, for example. Enhancements to increase reliability of the predictions are desired.

[0008] SUMMARY

[0009] The independent claims define the scope, and different embodiments are defined in dependent claims.

[0010] According to an aspect there is provided an apparatus comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: receive from a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; obtain measurement results of reference signals received by the apparatus; provide the measurement results as input to the first process; and adjust, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

[0011] In embodiments, the adjusting the multi-step process comprises: modifying the intermediate results output by the first process; and inputting modified intermediate results to the second process. In embodiments, the adjusting the multi-step process comprises: inputting to the second process also at least part of the quality assessment information or values determined from the quality assessment information.

[0012] In embodiments, the adjusting the multi-step process comprises: transmitting to the wireless network at least part of the quality assessment information; receiving from the wireless network an indication indicating whether or not to continue the multi-step process; and based on the indication indicating to continue, perform, according to the configuration, the second process.

[0013] In embodiments, the adjusting the multi-step process comprises: determining whether to perform, according to the configuration, the second process, or to transmit to the wireless network the measurement results provided as input to the first process; and based on a result of the determining, either perform the second process, or transmit to the wireless network the measurement results.

[0014] In embodiments, the adjusting the multi-step process comprises: determining whether to or not perform, according to the configuration, the second process; and based on a result of the determining, either perform or not to perform the second process.

[0015] In embodiments, the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive in the configuration an indication on quality assessment method to be used; and select a first process based on said indication.

[0016] In embodiments, the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: transmit to the wireless network a request for the configuration; and receive the configuration as a response to the request.

[0017] In embodiments, the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: transmit to the wireless network information at least on the predictions.

[0018] In embodiments, the predictions are mobility predictions.

[0019] In embodiments, the configuration indicates whether to use, for the quality assessment information, accuracy of intermediate results of the first process, and / or confidence estimate of intermediate results of the first process and / or confidence estimate of the predictions of the second process. According to an aspect there is provided an apparatus comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: transmit over a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus.

[0020] In embodiments, the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive from a second apparatus of the at least one second apparatus at least part of the quality assessment information; determine, based on the at least part of quality assessment information, whether or not the second apparatus is to continue the multi-step process; and transmit to the second apparatus an indication indicating whether or not to continue the multi-step process.

[0021] In embodiments, the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: determine, prior to transmitting, the configuration.

[0022] In embodiments, the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive from at least one of the at least one second apparatus a request for the configuration; and perform the determining and transmitting the configuration in response to the request being received.

[0023] According to an aspect there is provided a method comprising at least: receiving, by an apparatus, from a wireless network a configuration for a multi- step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; obtaining measurement results of reference signals received by the apparatus; providing the measurement results as input to the first process; and adjusting, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process. In embodiments, the adjusting the multi-step process comprises: modifying the intermediate results output by the first process; and inputting modified intermediate results to the second process.

[0024] In embodiments, the adjusting the multi-step process comprises: inputting to the second process also at least part of the quality assessment information or values determined from the quality assessment information.

[0025] In embodiments, the adjusting the multi-step process comprises: transmitting to the wireless network at least part of the quality assessment information; receiving from the wireless network an indication indicating whether or not to continue the multi-step process; and based on the indication indicating to continue, perform, according to the configuration, the second process.

[0026] In embodiments, the adjusting the multi-step process comprises: determining whether to perform, according to the configuration, the second process, or to transmit to the wireless network the measurement results provided as input to the first process; and based on a result of the determining, either perform the second process, or transmit to the wireless network the measurement results.

[0027] In embodiments, the adjusting the multi-step process comprises: determining whether to or not perform, according to the configuration, the second process; and based on a result of the determining, either perform or not to perform the second process.

[0028] In embodiments, the method further comprises at least: receiving in the configuration an indication on quality assessment method to be used; and selecting a first process based on said indication.

[0029] In embodiments, the method further comprises at least: transmitting to the wireless network a request for the configuration; and receiving the configuration as a response to the request.

[0030] In embodiments, the method further comprises at least: transmitting to the wireless network information at least on the predictions.

[0031] According to an aspect there is provided a method comprising at least: transmitting over a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus. In embodiments, the method further comprises at least: receiving from a second apparatus of the at least one second apparatus at least part of the quality assessment information; determining, based on the at least part of quality assessment information, whether or not the second apparatus is to continue the multi-step process; and transmitting to the second apparatus an indication indicating whether or not to continue the multi-step process.

[0032] In embodiments, the method further comprises: determining, prior to transmitting, the configuration.

[0033] In embodiments, the method further comprises: receiving from at least one of the at least one second apparatus a request for the configuration; and performing the determining and transmitting the configuration in response to the request being received.

[0034] According to an aspect there is provided a computer readable medium comprising instructions which, when executed by an apparatus, cause the apparatus to perform at least the following: receiving from a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; obtaining measurement results of reference signals received by the apparatus; providing the measurement results as input to the first process; and adjusting, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

[0035] According to an aspect, there is provided a computer readable medium comprising instructions which, when executed by an apparatus, cause the apparatus to perform at least the following: transmitting over a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus.

[0036] In embodiments, the computer readable medium further comprises instructions which, when executed by an apparatus, cause the apparatus to perform any embodiment, or embodiments of the method described above. In embodiments, the computer readable medium is a non-transitory computer readable medium.

[0037] According to an aspect there is provided a computer program comprising instructions, which, when executed by an apparatus, cause the apparatus to perform any aspect, embodiment, or embodiments of the method described above.

[0038] According to an aspect there is provided an apparatus comprising at least: means for receiving from a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; means for obtaining measurement results of reference signals received by the apparatus; means for providing the measurement results as input to the first process; and means for adjusting, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

[0039] In embodiments, the means for adjusting the multi-step process comprises: means for modifying the intermediate results output by the first process; and means for inputting modified intermediate results to the second process.

[0040] In embodiments, the means for adjusting the multi-step process comprises: means for inputting to the second process also at least part of the quality assessment information or values determined from the quality assessment information.

[0041] In embodiments, the means for adjusting the multi-step process comprises: means for transmitting to the wireless network at least part of the quality assessment information; means for receiving from the wireless network an indication indicating whether or not to continue the multi-step process; and means for performing, based on the indication indicating to continue, the second process according to the configuration.

[0042] In embodiments, the means for adjusting the multi-step process comprises: means for determining whether to perform, according to the configuration, the second process, or to transmit to the wireless network the measurement results provided as input to the first process; means for performing the second process, when a result of the means for determining is to perform the second process; and means for transmitting to the wireless network the measurement results, when a result of the means for determining is not to perform the second process.

[0043] In embodiments, the means for adjusting the multi-step process comprises: means for determining whether to or not perform, according to the configuration, the second process; means for performing the second process, when a result of the means for determining is to perform the second process; and means for not performing the second process, when a result of the means for determining is not to perform the second process.

[0044] In embodiments, the apparatus further comprises: means for receiving in the configuration an indication on quality assessment method to be used; and means for selecting a first process based on said indication.

[0045] In embodiments, the apparatus further comprises: means for transmitting to the wireless network a request for the configuration; and means for receiving the configuration as a response to the request.

[0046] In embodiments, the apparatus further comprises: means for transmitting to the wireless network information at least on the predictions.

[0047] In embodiments, the predictions are mobility predictions.

[0048] In embodiments, the configuration indicates whether to use, for the quality assessment information, accuracy of intermediate results of the first process, and / or confidence estimate of intermediate results of the first process and / or confidence estimate of the predictions of the second process.

[0049] According to an aspect there is provided an apparatus comprising at least: means for transmitting over a wireless network a configuration for a multi- step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus.

[0050] In embodiments, the apparatus further comprises: means for receiving from a second apparatus of the at least one second apparatus at least part of the quality assessment information; means for determining, based on the at least part of quality assessment information, whether or not the second apparatus is to continue the multi-step process; and means for transmitting to the second apparatus an indication indicating whether or not to continue the multi-step process.

[0051] In embodiments, the apparatus further comprises: means for determining, prior to transmitting, the configuration.

[0052] In embodiments, the apparatus further comprises: means for receiving from at least one of the at least one second apparatus a request for the configuration; wherein the means for determining and the means for transmitting are responsive to the request being received.

[0053] BRIEF DESCRIPTION OF DRAWINGS

[0054] Embodiments are described below, by way of example only, with reference to the accompanying drawings, in which

[0055] Fig. 1 illustrates an exemplified high-level network architecture;

[0056] Fig. 2 illustrates an example functionality;

[0057] Fig. 3 illustrates an example functionality;

[0058] Fig. 4 illustrates an example of information exchange;

[0059] Fig. 5 illustrates an example of information exchange;

[0060] Fig. 6 illustrates an example of information exchange;

[0061] Fig. 7 illustrates an example of information exchange;

[0062] Fig. 8 illustrates an example of a multi-step process;

[0063] Fig. 9 illustrates an example of an adjusted multi-step process;

[0064] Fig. 10 illustrates an example of an adjusted multi-step process;

[0065] Fig. 11 illustrates an example of an adjusted multi-step process;

[0066] Fig. 12 illustrates an example of an adjusted multi-step process;

[0067] Fig. 13 illustrates an example of an adjusted multi-step process;

[0068] Fig. 14 illustrates an example of training arrangement;

[0069] Fig. 15 is a schematic block diagram;

[0070] Fig. 16 is a schematic block diagram;

[0071] Fig. 17 is a schematic block diagram; and

[0072] Fig. 18 is a schematic block diagram.

[0073] DETAILED DESCRIPTION OF SOME EMBODIMENTS

[0074] The following embodiments are only presented as examples. Although the specification may refer to “an”, “one”, or “some” embodiments] and / or examplefs] in several locations, this does not necessarily mean that each such reference is to the same embodiments] or examplefs], or that a particular feature only applies to a single embodiment and / or single example. Single features of different embodiments and / or examples may also be combined to provide other embodiments and / or examples. Furthermore, words “comprising” and “including” should be understood as not limiting the described embodiments to consist of only those features that have been mentioned and such embodiments may contain also features / structures that have not been specifically mentioned. Further, although terms including ordinal numbers, such as “first”, “second”, etc., may be used for describing various elements, the elements are not restricted by the terms. The terms are used merely for the purpose of distinguishing an element from other elements. For example, a first element could be termed an element or a second element, and similarly, a second element could be also termed a first element or an element without departing from the scope of the present disclosure.

[0075] 5G-Advanced, and beyond future wireless networks aim to support a large variety of services, use cases and industrial verticals, for example unmanned mobility with fully autonomous connected vehicles, other vehicle-to-everything (V2X) services, or smart environment, e.g. smart industry, smart power grid, or smart city, just to name few examples. To provide variety of services with different requirements, such as enhanced mobile broadband, ultra-reliable low latency communication, massive machine type communication, wireless networks are envisaged to adopt network slicing, flexible decentralized and / or distributed computing systems and ubiquitous computing, with local spectrum licensing, spectrum sharing, infrastructure sharing, and intelligent automated management underpinned by mobile edge computing, artificial intelligence, for example machine learning, based tools, cloudification and blockchain technologies. For example, in the network slicing multiple independent and dedicated network slice instances may be created within the same infrastructure to run services that have different requirements on latency, reliability, throughput and mobility.

[0076] 6G (sixth generation) networks are expected to adopt flexible decentralized and / or distributed computing systems and architecture and ubiquitous computing, with local spectrum licensing, spectrum sharing, infrastructure sharing, and intelligent automated management underpinned by mobile edge computing, artificial intelligence, short-packet communication and blockchain technologies. Key features of 6G will include intelligent connected management and control functions, programmability, integrated sensing and communication, reduction of energy footprint, trustworthy infrastructure, scalability and affordability. In addition to these, 6G is also targeting new use cases covering the integration of localization and sensing capabilities into system definition to unifying user experience across physical and digital worlds.

[0077] Fig. 1 illustrates an exemplified extremely high-level network architecture only showing some functional entities, all being logical units, whose implementation may differ from what is shown. The connections shown in Fig. 1 are logical connections; the actual physical connections may be different.

[0078] The system 100 depicted in Fig. 1 is based on the 5G system (fifth generation system). The examples are described herein using principles and terminology of 5G without limiting the examples, and the terminology used to the 5G. A person skilled in the art may apply the solutions and examples to other communication systems, for example beyond 5G, e.g. 6G, 7G, provided with necessary properties.

[0079] Referring to Fig. 1, a wireless network 100, or a system comprising wireless networks, comprises device components 101 for device functionalities in device domain, access network components 102 for access network functionalities in access network domain, core network components 103 for core network functionalities in core network domain, and data network components 104 for data network functionalities in data network domain.

[0080] A device component 101 may be any electrical device, or apparatus, connectable to an access network and configurable to be in a wireless connection on one or more communication channels, including one or more control channels, with an access network component 102 providing a cell, for example. The physical link from the device component 101 to the access network component 102 towards the core network component 103 is called an uplink or a reverse link and the physical link to the device component is called a downlink or a forward link. By way of example rather than limitation, the device component 101 may referred to as a mobile apparatus, a terminal device, a communication device, a user equipment (UE), a subscriber station CSS}, a portable subscriber station, a mobile station (MS), or an access terminal (AT). A non-limiting lists of examples of the device component 101, or what the device component 101 may comprise or be comprised in, include a mobile phone, a cellular phone, a smart phone, a voice over internet protocol (VoIP) phone, a wireless local loop phone, a device using a wireless modem, a portable computer, a desktop computer, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), a smart device, a multimedia device, an image capture terminal device, such as a digital camera, a gaming terminal device, a music storage and playback appliance, a drone, a vehicle, an automated guided vehicle, an autonomous connected vehicle, a vehicle-mounted wireless terminal device, a wireless endpoint, an internet of things device, an industrial internet of things device, a device operating in an industrial and / or an automated processing chain contexts, a consumer electronics device, a consumer internet of things device, a mobile robot, a mobile robot arm, a sensor, a surveillance camera, an eHealth related device, a medical monitoring device, a medical device, for example for remote surgery, a wearable device, such as a smart watch, a smart ring, a head-mounted display (HMD), an on-person device, etc. The device component may also be part of a group of device components seen as one device component, i.e. one mobile apparatus, by the wireless network.

[0081] An access network domain may be based on any kind of an access network, such as a cellular access network, for example 5G network, 5G-Advanced network, 6G network, etc., a non-terrestrial network, a legacy cellular radio access network, for example 4G or older generation network, or a non-cellular access network, for example a wireless local area network, or any combination thereof. To provide the wireless access, the access network comprises access network components 102, such as access network apparatuses, or access devices. An access device component 102 may provide one or more cells, possibly with different cell accessibility per a cell, but a cell is provided by one access device. However, there may be overlapping cells, for example a macro cell provided by an access device operating in co-operation of access nodes providing smaller cells, such as micro-, femto- or picocells, which overlap at least partly within the macro cell. There are a wide variety of access network components 102. A non-limiting lists of examples of the access network component 102, or what the access network component 102 may comprise or be comprised in, include different types of base stations, such as eNBs, gNBs, split gNBs, transmission-reception points, network-controlled repeaters, nodes operationally coupled to one or more remote radio heads, satellites, donor nodes in integrated access and backhaul (LAB), fixed 1AB nodes, mobile 1AB nodes mounted on vehicles, for example, etc. At least some of the apparatuses in the access network may provide an abstraction platform to separate abstractions of network functions from the processing hardware.

[0082] Further, it should be noted that some of the components may be multidomain components. For example a device component 101 may also provide services to other device components, i.e. operate also as an access network component 102, for example be a relay node, or a mobile 1AB node, or a mobile termination part in an 1AB node. Hence, herein term mobile apparatus is used for device components, or device component functionality in a multi-domain component and term access network apparatus is used for access network components or access network component functionality in a multi-domain component.

[0083] The core network components 103 form one or more core networks. A core network may be based on a non-standalone core network, for example an LTE- based network, or a standalone access network, for example a 5G core network. However, it should be appreciated that the core network, and the core network components 103, may use any technology that enable network services to be delivered between devices and data networks.

[0084] A data network may be any network, like the internet, an intranet, a wide area network, etc. Different remote monitoring and / or data collection services for different use cases may be reached via the data network and the data network components 104.

[0085] It is envisaged that multi-step processes will be used to obtain different predictions, for example mobility predictions, in different use scenarios. Mobility predictions may be, for example, predictions for future mobility events, future strongest beams or cells, ping-pongs, radio coverage failure, etc. A multi-step process comprises at least a first process and a subsequent second process that are performed separately and the input to the second process is based on output of the first process, the first process outputting at least intermediate results. The intermediate results may be called intermediate predictions. The second process outputs the predictions. The first process may be a generic process and the second process a more specialized process, more easily configurable by a network. Hence, the multi-step process provides flexibility, and since the processes are separate processes, the designing of the processes is simpler than designing one process for a similar result. The processes in a multi-step process may be performed by one apparatus, for example, a mobile apparatus may perform the first process and the second process, or different apparatuses may perform different processes, for example the mobile apparatus may perform the first process and the access network apparatus the second process. The first process may be a machine learning model, or another process (non-machine learning based process) outputting intermediate results. The second process may be any process, including non-machine learning based process and a machine learning model, which may be different from the machine learning model for the first process, if the first process is implemented using the machine learning model. A use case example includes the first process outputting reference signal received power values and the second process predicting so called mobility ping-pong events from the reference signal received power values. Other mobility events that may be predicted include future handovers, handover failure, radio link failure, etc.

[0086] For example, input to the first process may have varying sample distributions causing varying uncertainty to the predictions. To increase the accuracy of predictions, at least for some future events, for example for mobility events near a cell edge, the multi-step process may be adjusted, as will be described in more detail below, based on quality assessment information of the intermediate results. The quality assessment information may indicate an accuracy for the first process, for example obtained based on predictions preceding the current prediction. The quality assessment information may also be a confidence estimate, which may refer to the intermediate results of the first process, and / or events relating to the predictions of the second process, for example before the events have been performed (carried out).

[0087] Fig. 2 is a flow chart illustrating an example functionality of an apparatus (a mobile apparatus) relating to a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions.

[0088] Referring to Fig. 2, the apparatus receives (block 201) from a wireless network a configuration for the multi-step process. The configuration may, for example, comprise an indication on quality assessment method to be used. For example, the configuration may indicate whether to use, for the quality assessment information, accuracy of intermediate results of the first process, and / or confidence estimate of intermediate results of the first process and / or confidence estimate of the predictions of the second process. Based on said indication, a first process to be used in the multi-step process may be selected. For example, the mobile apparatus may have two or more possible first processes for the multi-step process, or selection of the first process includes selection of the multi-process amongst two or more different multi-processes. For example, there may be a multi- step process not performing quality assessment at all, for example such as described below with Fig. 8, and then one or more multi-step processes implementing different quality assessments. The indication on the quality assessment method may be used also in an additional step of assessing quality.

[0089] After receiving the configuration, the apparatus applies the configuration. In other words, when the apparatus obtains (block 202) measurement results of reference signals received by the apparatus, the measurement results are provided (203) as input to the first process, and the apparatus adjusts (block 204), according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

[0090] The adjusting may comprise modifying the intermediate results output by the first process, and inputting modified intermediate results to the second process. For example, instead of using an intermediate result of the first step, or a median calculated from intermediate results, or from a predetermined number of intermediate results, an upper or lower value of a percentile determined based on the configuration and / or quality assessment information, in a confidence interval for the first process may be used as a modified intermediate result. This enables erring on the side of caution, but still utilizing the intermediate result. For example, if very reliable predictions are required, the quality assessment information that is used by the second process may be that 90%-tile interval of RSRP instead of the median, erring to the side of caution. On the other hand, if high detection rate is required and the cost of false predictions is low, the second process may use another percentile intervaL-

[0091] Alternatively, or in addition to the above, the adjusting the multi-step process may comprise inputting to the second process also at least part of the quality assessment information or values determined from the quality assessment information.

[0092] Alternatively, or in addition to the above, the adjusting the multi-step process may comprise transmitting to the wireless network at least part of the quality assessment information, receiving from the wireless network an indication indicating whether or not to continue the multi-step process. If the indication received indicates to continue the multi-step process, performing the second process, according to the configuration received earlier. Accordingly, the apparatus performs the second process based on the indication indicating to continue the multi-step process. If the indication received indicates not to continue the multi- step process, the second process is not performed. Accordingly, the apparatus does not perform the second process based on the indication indicating not to continue the multi-step process. In other words, the apparatus determines, based on the indication, the continuation of the multi-step process.

[0093] Alternatively, or in addition to the above, the adjusting the multi-step process may comprise determining whether to perform the second process, or to transmit to the wireless network the measurement results provided as input to the first process. When the determining result is to perform the second process, the second process is performed according to the configuration. When the determining results not to perform the second process, the measurement results are transmitted to the wireless network. Hence, no predictions are obtained, and the second process is not performed in vain, for example when there is a high uncertainty. Accordingly, the apparatus performs the second process, according to the configuration, based on the determining result being to perform the second process. Accordingly, based on the determining result being to not perform the second process, the apparatus does not perform the second process and instead transmits the measurement results to the wireless network. In other words, the apparatus determines, based on the determining result, the continuation of the multi-step process.

[0094] Alternatively, or in addition to the above, the adjusting the multi-step process may comprise determining whether to or not perform, according to the configuration, the second process. When a result of the determining is to perform the second process, it will be performed. When the result is not to perform, no second process is performed, and no predictions are obtained. Also herein, the second process is not performed in vain, for example when there is a high uncertainty. Accordingly, the apparatus performs the second process, according to the configuration, based on determining to perform the second process. Accordingly, based on determining to not perform the second process, the apparatus does not perform the second process and does not obtain predictions. In other words, the apparatus determines, based on the configuration, the continuation of the multi-step process.

[0095] Alternatively, or in addition to the above, the adjusting the multi-step process may comprise modifying postprocessing parameterization of the second process based on the quality assessment information. For example, the second process may perform cell-switch prediction based on predicted reference signal received power (RSRP), output by the first process as an intermediate result. For a low quality assessment value, the cell-switch prediction (in this case postprocessing of the intermediate result of the first process) may apply parameterization which is more robust. For instance when the intermediate results of the first process are analyzed for cell-switches, the analysis / postprocessing may use larger neighbour-better-than-serving offsets or larger time to trigger (TTT) values .

[0096] Fig. 3 is a flow chart illustrating an example functionality of an apparatus (an access network apparatus in a wireless network) relating to the multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions. The example functionality provides the wireless network, or the wireless system comprising the wireless network, capability to configure and reconfigure quality assessment and how the multi-step process is adjusted.

[0097] Referring to Fig. 3, the apparatus transmits (block 301) over the wireless network a configuration for the multi-step process to at least one second apparatus. The transmission is a downlink transmission, and a second apparatus may be a mobile apparatus. Examples what the configuration may comprise are given above with Fig. 2, for example.

[0098] The transmission of the configuration to a second apparatus (a mobile apparatus) may be triggered by the second apparatus requesting the configuration and / or when the second apparatus connects to a cell provided or controlled by the access network apparatus, for example.

[0099] Fig. 4, Fig. 5, Fig. 6 and Fig. 7 illustrate different examples of information exchange between a mobile apparatus and an access network apparatus, i.e. AN apparatus relating to the multi-step process. In the examples of Fig. 4, Fig. 5 and Fig. 7 the mobile apparatus is hosting the multi-step process. The advantage is that less signaling is required between mobile apparatus and the wireless network (the AN apparatus). Further advantage is that decisions can be made faster, since no delay is caused by the signaling, enabling frequent running of the adjustable multi- step process. In the example of Fig. 6, the wireless network is involved in a decision making process. When the wireless network is involved, less computing capacity of the mobile apparatus is used, and hence computing capacity requirements for the mobile apparatus may be lower, and the wireless network has more control on future events. Referring to Fig. 4, the mobile apparatus transmits (message 4-1) to the AN apparatus (to the wireless network) a request for the configuration. The request may be an indirect request, for example it may be a capability information message, or corresponding signaling, which indicates at least some capabilities of the mobile apparatus. The capability information may comprise, for example as an information element, multi-step processing capability or multi-step adjusting capability of the mobile apparatus. For example, information indicating different quality assessment functionalities the mobile apparatus has and / or other alternatives for the multi-step process(es) the mobile apparatus has, may be transmitted in message 4-1. The multi-step processing capability may be given as a prediction processing capability or as indication of such capability. The request (message 4-1) may be also a specific request for the configuration, for example a multi-step quality configuration message comprising at least multi-step processing capability information. Depending on an implementation, message 4-1 may be transmitted one or more times, for example when the mobile apparatus connects to the network, and / or when the mobile apparatus enters a mobility state for which there is a multi-step process outputting predictions usable for mobility events in the mobility state. For example, message 4-1 may be transmitted when a speed of the mobile apparatus exceeds a preset threshold. Further, it should be appreciated that even though not illustrated in Fig. 4, the request, or capability information relating to the multi-step process, may be relayed between access network apparatuses as part of the mobile apparatus context.

[0100] In the example of Fig. 4, the AN apparatus determines (block 4-2), based at least on the received capability information, the configuration. The determining in block 4-2 may be performed in response to the request (message 4-1) being received, or at a later stage, for example when the wireless network, or the AN apparatus, determines that prediction processing capability, or multi-step processing capability of the mobile apparatus, would allow making predictions, for example mobility predictions, in current channel conditions. Different examples of the configuration are given above with Fig. 2. Further non-limiting examples of what the configuration may indicate include one or more of the following:

[0101] • Validity period of the accuracy, e.g., how often the accuracy is to be updated

[0102] • Whether or not accuracy is to be averaged over preceding measurement result. • If the accuracy is to be averaged, averaging period, i.e. how long the accuracy is to be averaged

[0103] • Whether or not use confidence-based quality assessment

[0104] • If the confidence-based quality assessment is to be used, whether the confidence of the intermediate results of the first process and / or the confidence of predictions of the second process is to be taken into account, for example as will be explained below.

[0105] • The way the quality assessment information is to adjust (modify) the multi-step process, for example the way to adjust the input to the second process, or parameterization of the second process, or when the perform the second process

[0106] • What to report to the network, for example to include or not at least part of quality assessment information, to report or not at least part of the intermediate results and / or input to the first process.

[0107] The configuration is then transmitted (message 4-3) over the wireless network to the mobile apparatus. In other words, the mobile apparatus may be configured with the type of quality assessment to be performed and / or how to adjust the multi-step process, or which one of possible processes are to be performed and / or what to report.

[0108] The wireless network, for example, the AN apparatus, and other AN apparatuses, transmits reference signals (depicted by message 4-4), and the mobile apparatus obtains (block 4-5) measurement results of the reference signals. The reference signals may be positioning reference signals and / or channel state information reference signals, or channel status information reference signals, for example. The measurement results may comprise reference signal received power, reference signal received path power, reference signal received quality, channel impuls response, cell identifiers, beam identifiers, and / or angle of arrival, for example.

[0109] In the illustrated example of Fig. 4, the multi-step process is configured and will be performed. This means that the measurement results, or at least part of the measurement results are input in block 4-6 to the first process and in the illustrated example of Fig. 4, an adjusted second process, according to the configuration received in message 4-3, is performed in block 4-6. For example, a quality assessment may be performed in block 4-6, for example as described above with Fig. 2, or as will be described below. In the illustrated example of Fig. 4, it is assumed that two reports (message 4-7 and message 4-8) are transmitted, according to the configuration, to the AN apparatus. For example, message 4-7 may comprise quality assessment information output by the first process, and / or inputs to the second process, and / or parameters used in the second process, and message 4-8 may be a prediction report comprising one or more predictions output by the second process. In another example, the mobile apparatus transmits, after block 4-6, information at least on the predictions, for example mobility predictions, to the wireless network. The reporting of the quality assessment information serves the purpose of enabling the wireless network to monitor the scenario. For example, if the mobile apparatus consistently reports low quality, and the multi-step process is implemented using models, the wireless network may trigger fine-tuning of the models, or one of the models.

[0110] Block 4-6, and related reporting may be performed continuously, or at certain intervals, or when specifically triggered, for example when the mobile apparatus is at a cell border or a handover or a cell switch is imminent. For example, an occurrence of a configured mobility event, such as A3 event, may trigger the functionality in block 4-6.

[0111] Messages 4-1 and 4-3 may be part of radio resource control (RRC) signaling. Messages 4-7 and 4-8 may also be part of the radio resource control (RRC) signaling, or message 4-7 and / or message 4-8 may be part of a lower level signaling, for example included in medium access control (MAC) reports.

[0112] Fig. 5 illustrates information exchange according to an implementation or scenario wherein no specific request is transmitted from the mobile apparatus to trigger transmission of the configuration, and the mobile apparatus is configured to report the predictions.

[0113] Referring to Fig. 5, the access network transmits (message 5-1) the configuration. Message 5-1 corresponds to message 4-3, described in detail above. The transmission may be triggered by the mobile apparatus connecting to a cell provided or controlled by the AN apparatus, for example, or a service being triggered for the mobile apparatus, the service requiring predictions.

[0114] The wireless network, for example, the AN apparatus, and other AN apparatuses, transmits reference signals (depicted by message 5-2, corresponding to message 4-4 in Fig. 4), and the mobile apparatus obtains (block 5-3, corresponding to block 4-5 in Fig. 4) measurement results of the reference signals. In the illustrated example of Fig. 5, the multi-step process is configured and will be performed. This means that the measurement results, or at least part of the measurement results are input in block 5-4 to the first process and in the illustrated example of Fig. 5, an adjusted second process, according to the configuration received in message 5-1, is performed in block 5-4. Block 5-4 corresponds to block 4-6, described in detail above.

[0115] In the illustrated example of Fig. 5, it is assumed that one report (message 5-5) is transmitted, message 5-5 corresponding to message 4-8 described above.

[0116] In still further implementations, the example of Fig. 5 contains further that prior to transmitting the configuration (message 5-1), the AN apparatus determines, for example as described above in block 4-2, using e.g. the mobile apparatus context, the configuration, or transmits either in message 5-5 a report corresponding to the report in message 4-7 described above or in addition to message 5-5 also a report corresponding the report in message 4-7.

[0117] In the below examples of Fig. 6 and Fig. 7 the information exchange illustrated starts when the AN apparatus transmits the configuration. It should be appreciated that information exchange corresponding to message 4-1 and block 4-2 may have taken place prior the illustrated information exchange.

[0118] Referring to Fig. 6, the access network transmits (message 6-1) the configuration. Message 6-1 corresponds to message 5-1 and message 4-3, described in detail above.

[0119] The wireless network, for example, the AN apparatus, and other AN apparatuses, transmits reference signals (depicted by message 6-2, corresponding to message 4-4 in Fig. 4), and the mobile apparatus obtains (block 6-3, corresponding to block 4-5 in Fig. 4) measurement results of the reference signals.

[0120] In the illustrated example of Fig. 6, the multi-step process is configured and will be performed. This means that the measurement results, or at least part of the measurement results are input in block 6-4 to the first process. However, in the example of Fig. 6, at least part of the quality assessment information, which the first process outputs is transmitted (message 6-5) from the mobile apparatus to the AN apparatus, according to the configuration received in message 6-1. Different examples of quality assessment information are described above.

[0121] The AN apparatus determines (block 6-6), based on the at least part of quality assessment information, whether or not the mobile apparatus is to continue the multi-step process. In other words, the wireless network side controls, whether the second process will be performed. This has the advantage that the wireless network can implement a decision logic and more flexibly adjust usage of the second process. Depending on an implementation, the determining in block 6-6 may include to determine a validity time for the decision, or a preset validity time for the decision may be used, or no validity time is used, meaning that determining will be implemented in a more frequent manner. The result of the determination is then transmitted (message 6-7) to the mobile apparatus. The indication may be ACK indicating that the quality is good enough to continue to the second process, or NACK not to continue to the second process. If the result indicates that the second process is not to be performed, it will not be performed. Further, if the result is associated with the validity time, the mobile apparatus may be configured not to perform the first process within the validity time, or at least not to transmit message 6-5 and not to perform the second process until the validity time has expired. Messages 6-5 and 6-7 are part of adjusting the multi-step process at the mobile apparatus.

[0122] If the result indicates, as is assumed in the example illustrated in Fig. 6, that the second process is to be performed, the second process is performed in block 6-8. The second process may be adjusted, as described above, or performed without any further adjustments, since the multi-step process is adjusted by the first process outputting the quality assessment information and by continuing to the second process if the wireless network so determines. Further, if the result is associated with the validity time, the mobile apparatus may be configured not to not to transmit message 6-5 and perform the second process until the validity time has expired.

[0123] In the illustrated example of Fig. 6, when the second process is performed (block 6-8) the predictions are signaled (message 6-9, corresponding to message 4-8) to the AN apparatus.

[0124] Messages 6-5 and 6-7 may be part of the radio resource control (RRC) signaling, or message 6-5 and / or message 6-7 may be part of a lower level signaling, for example included in medium access control (MAC) reports.

[0125] Referring to Fig. 7, the access network transmits (message 7-1) the configuration. Message 7-1 corresponds to message 5-1 and message 4-3, described in detail above. In the example of Fig. 7, it is assumed that the configuration indicates a threshold to be used when determining whether or not to continue to the second process. The wireless network, for example, the AN apparatus, and other AN apparatuses, transmits reference signals (depicted by message 7-2, corresponding to message 4-4 in Fig. 4), and the mobile apparatus obtains (block 7-3, corresponding to block 4-5 in Fig. 4) measurement results of the reference signals.

[0126] In the illustrated example of Fig. 7, the multi-step process is configured and will be performed. This means that the measurement results, or at least part of the measurement results are input in block 7-4 to the first process, which is adjusted to use the quality assessment information, or part of it, for a comparison to determine whether or not to perform the second process. For example, a quality assessment value, qv, provided in the quality assessment information or calculated using the quality assessment information of the first process, may be compared to a predefined threshold, th, which may be received in the configuration. In the illustrated example, it is assumed that when the quality assessment value is below the threshold, i.e. qv < th, the quality is not good enough, the second process is not performed. This is illustrated in block 7-4. Depending on an implementation and / or the received configuration, when the second process is not performed, the mobile apparatus may transmit (message 7-5 with dashed line indicating optionality) to the AN apparatus the input to the first process, i.e. at least part of the measurement results and / or at least part of the quality assessment information which the first process outputs. The AN apparatus may then process (block 7-6) the information. As can be seen from the example, the quality assessment information serves to disable the second process. Further, message 7-5 may be part of the adjustment of the multi-step process at the mobile apparatus.

[0127] In the illustrated example of Figure 7, when a next occurrence when the wireless network, for example, the AN apparatus, and other AN apparatuses, transmits reference signals (depicted by message 7-2a, corresponding to message 4-4 in Fig. 4), and the mobile apparatus obtains (block 7-3a, corresponding to block 4-5 in Fig. 4) measurement results of the reference signals, and at least part of the measurement results are input in block 7-4a, corresponding to block 7-4, to the first process, the quality assessment value is above the threshold, i.e. qv > th, the quality is good enough, the second process is performed in block 7-7, corresponding to block 6-8 described above. Further, in the illustrated example of Fig. 7 the predictions are signaled (message 7-8, corresponding to message 4-8) to the AN apparatus. The quality assessment information, may also be signaled to the AN apparatus. Hence, in the example of Fig. 7, the second process is not performed in vain. Further, it is possible to provide the wireless network with information (at least part of the measurement result) based on which the wireless network may determine at least that no predictions will be received during that predication occurrence, and determine how to continue.

[0128] Fig. 8 to Fig. 13 illustrate examples of multi-step process implemented using a hierarchical machine learning based model comprising step 1 model performing the first process and step 2 model performing the second process. As machine learning is comprised in artificial intelligence, it should be appreciated that herein the term machine learning covers any artificial intelligence field based model. In the illustrated examples, the process is described, for the sake of clarity, assuming that predictions are mobility predictions, and the intermediate results are future predicted reference signal received power, RSRP, values, without limiting the examples to mobility predictions and RSRP values.

[0129] Fig. 8 illustrates an example of a multi-step model 810 which is not adjusted. The model may be adjusted to implement the example of Fig. 6 or Fig. 7 by adjusting the step 1 model 811 to output the quality assessment information, and to input intermediate results 802 to the step 2 model 812 when the step 2 model is to be performed. However, the adjustment is not illustrated in Fig. 8.

[0130] Referring to Fig. 8, the step 1 model 811 to perform prediction of a RSRP sequence, output as intermediate result 802, may comprise one or more convolutional or recurrent neural networks or long-term short-term (LSTM) networks. For instance, the step 1 model 811 may comprise 2 layers of Conv2D (2D convolution layer) networks or LSTM networks, followed by 2 layers of dense (densely connected neurons) networks. The step 1 model 811 may use as input 801 a sequence of past RSRP measurements of a beam, for several beams. The sequence of past measurements may comprise, for example, 20 samples, where samples may have been collected every 20 msec. The step 1 model 811 may use samples collected on 6 beams as input 801, for example. Thus, there may be 20*6 samples spanning a past window of 400 msec used as input 801. The output 802 of the step 1 model 811 may be a sequence of RSRP values, wherein for each input beam sequence there may be one output beam sequence. Hence, the step 1 model 811 can be considered to perform a type of regressive prediction. It should be appreciated that the step 1 model 811 may use any type of sequence prediction to output the intermediate results 802. The step 2 model 812, or postprocessing step 2, may be implemented as a parser, which analyses whether the RSRP predictions 802 output by the step 1 model exhibit mobility events, for example A3 events, and outputs 803 predictions of the events. For example, the step 2 model 812 may be a classifier, which may be used to determine future mobility events or actions to be taken based on the RSRP predictions 802 as intermediate results.

[0131] A quality assessment Information relating to output 802 from the step 1 model may be a confidence interval, which demonstrates a range of values that is likely to contain the true value of a population parameter based on some samples of input data 801. Confidence intervals may be computed using resampling methods (e.g., bootstrapping), cross validation, Bayesian approach, etc.

[0132] When the step 1 model 811 comprises regression models, the quality assessment information relating to the output 802 may also be obtained by calculating a mean average error (MAE) over the validation dataset. The validation dataset is usually separate from a training dataset, and the validation dataset is usually for validating a trained model, for example for tuning hyperparameters of the trained model and / or to test whether the trained model is well generalized for different events. The validation dataset may be collected in a period preceding a prediction, or it may have be collected when the training dataset is collected, or a retraining dataset is collected. Machine learning based models, such as mixture density networks, may be trained to output the quality assessment information relating to the output 802, wherein the quality assessment information may comprise, or be based on, all parameters of a distribution, like the mean and standard deviation, of all the distributions mixed in the general distribution, given the specific input features. For example, these machine learning based models may be used to estimate the confidence intervals in inference of regression models.

[0133] Fig. 9 illustrates an example how the multi-step model depicted in Fig. 8 may be adjusted to be an adjusted multi-step model 910. In the illustrated example of Fig. 9 it is assumed that the input 801 to the step 1 model 911 remains the same as in Fig. 8, and the output 802 of intermediate results also remains the same as in Fig. 8. However, the first step model 911 is adjusted to output 901 the quality assessment information and the step 2 model 912 is adjusted to receive both outputs 802 and 901 and output at least predictions 903 obtained based on the two inputs. In an implementation, the step 2 model 912 may be adjusted to output 901 also the quality assessment information, or at least part of it. The quality assessment information, depicted by one output-input 901, may comprise one or more outputs-inputs. For example, it may comprise confidence intervals, for example upper and lower confidence intervals.

[0134] For example, the step 1 model 911 may be a predictive regression model taking a time series of Tin latest RSRP measurements of the N strongest beams as an input. Using the input it is predicting the future RSRP measurements for the N beams for a time window of Tout samples as the intermediate result 802. In addition to the point values 802 representing the expected mean RSRP being provided to the step 2 model 912, the step 1 model 911 is also predicting the distribution of the RSRP measurements, and outputs, as quality assessment information, the distribution of the RSRP measurements to be input 901 to the step 2 model 912, for example in form of selected confidence interval upper and lower bounds. These confidence intervals also serve as a quality assessment. The step 2 model 912 may be a classifier, which is determining future mobility events or actions to be taken based on the RSRP prediction. Now, in addition to the predicted mean point value, the classifier (the step 2 model 912) is trained to use the input 901, i.e. the confidence interval bounds, to make more robust decisions. If the expected quality is very low the output 903 may for example not recommend any actions. Alternatively, the step 2 model 912 may apply thresholds to the confidence intervals. For example, to determine that a cell will remain strong enough to be connected to, the 2 step model 912 may be configured, for example by the configuration comprising corresponding values, to require that the 90 % confidence interval of the intermediate result 802 of the step 1 model is above the radio link failure threshold, instead of using a predicted mean value.

[0135] The model 910 in Fig. 9 may be adjusted to implement the example of Fig. 6 or Fig. 7 by using the quality assessment information 901 as described with Fig. 6 or Fig.7, and to input intermediate results 802 and the quality assessment information 901 to the step 2 model 912 when the step 2 model is to be performed. However, the adjustment is not illustrated in Fig. 9.

[0136] Fig. 10 illustrates another example how the multi-step model depicted in Fig. 8 may be adjusted to be an adjusted multi-step model 1010, resulting with the same input 801 to same output predictions 903 as the multi-step model depicted in Fig. 9 .

[0137] Referring to Fig. 10, the outputs 802 and 901 from the step 1 model 911 are modified by a modifier (mod) 1011 that is configured to modify the output 802, based on the output 901, so that modified intermediate results are input to the step 2 model 812’. The step 2 model 1012 may be the step 2 model 812, that may, or may not, have been trained to use also the modified intermediate results. The modifier may be integrated to be part of output layer of the step 1 model, for example to be an output selection algorithm.

[0138] The model 1010 in Fig. 10 may be adjusted to implement the example of Fig. 6 or Fig. 7 by using the quality assessment information 901 as described with Fig. 6 or Fig.7, and to input intermediate results 802 and the quality assessment information 901 to the modifier 1011 when the step 2 model is to be performed. However, the adjustment is not illustrated in Fig. 10.

[0139] Fig. 11 illustrates a further example how the multi-step model depicted in Fig. 8 may be adjusted to be an adjusted multi-step model 1110. In the illustrated example of Fig. 11, the quality assessment information 901 output by the step 1 model 911 is used in parameter selection (par.sel.) portion 1113 of the step 2 model 1112. In other words, parameterization of the step 2 model 1112 processing of intermediate results 802 of the step 1 model 911 is influenced by quality assessment information output by the step 1 model 911 while the intermediate results 802 are fed without modification.

[0140] For example, for different quality levels produced by the step 1 model 911 different parameterization is applied in the step 2 model. A non-limiting example includes thresholds which are applied to determine whether one cell is becoming stronger than another, or a time duration (time to trigger, TTT] that needs to be satisfied for one cell to be stronger than another before a cell switch is predicted.

[0141] The model 1110 in Fig. 11 may be adjusted to implement the example of Fig. 6 or Fig. 7 by using the quality assessment information 901 as described with Fig. 6 or Fig.7, and to input when the step 2 model is to be performed intermediate results 802 to the step 2 model 1112 and the quality assessment information 901 to the parameter selection portion 1113. However, the adjustment is not illustrated in Fig. 11.

[0142] Fig. 12 illustrates a further example how the multi-step model depicted in Fig. 8 may be adjusted to be an adjusted multi-step model 1210. In the illustrated example of Fig. 12, the quality assessment information is utilizing accuracy of previous intermediate results, such as RSRP predictions. As in the example of Fig. 11, also in the example of Fig. 12, the quality assessment information 901 output by the step 1 model 1211 is used in parameter selection (par.sel.) portion 1113 of the step 2 model 1112. Referring to Fig. 12, the step 1 model 1211 comprises a model 811’, which may be the step 1 model 811 described with Figure 8, which outputs the intermediate results 802. The step 1 model 1211 further comprises a delay functionality 1221 and an accuracy (acc.) computation functionality 1222. The delay functionality 1221 may implement a predicting horizon delay model. The delay functionality 1221 receives as input the intermediate results 802, which are predictions of future RSRP values which should be obtained at time tl, and feeds 1202 at the time tl the intermediate results to the accuracy computation functionality 1222. The accuracy computation functionality 1222 also receives as input the measurement results 801, e.g., the measurement results obtained at time tl, and hence can compare the accuracy of the intermediate results to measured results. In other words, predictions of future RSRP values at time tl can be compared to RSRP values obtained at time tl. The accuracy computation functionality 1222 outputs the comparison result, i.e. accuracy information, as the quality assessment information 901 to the parameter selection portion 1113 of the step 2 model 1112, to influence to the parameterization, as described above.

[0143] The model 1210 in Fig. 12 may be adjusted to implement the example of Fig. 6 or Fig. 7 by using the quality assessment information 901 as described with Fig. 6 or Fig.7, and to input when the step 2 model is to be performed intermediate results 802 to the step 2 model 1112 and the quality assessment information 901 to the parameter selection portion 1113. However, the adjustment is not illustrated in Fig. 12.

[0144] Fig. 13 illustrates a further example how the multi-step model depicted in Fig. 8 maybe adjusted to be an adjusted multi-step model 1310. In the illustrated example of Fig. 13, the quality assessment information is confidence information of expected output 903 determined prior corresponding input to the step 2 model 1112. As in the example of Fig. 11, also in the example of Fig. 13, the quality assessment information 901 output is used in parameter selection (par.sel.) portion 1113 of the step 2 model 1112.

[0145] Referring to Fig. 13, the step 1 model 1311 comprises a model 811’, which may be the step 1 model 811 described with Figure 8, which outputs the intermediate results 802. The step 1 model 1311 further comprises a confidence (conf.) model 1312. The confidence model 1311 receives as inputs the intermediate results 802 and the measurement results 801 and outputs confidence information, e.g. confidence intervals, or confidence interval estimations, as the quality assessment information 901 to the parameter selection portion 1113 of the step 2 model 1112, to influence to the parameterization, as described above. The confidence model may be a mixture density network based model. The confidence model may be a classification model that uses as output of its neural network a softmax layer, which is a probability distribution over possible classes or categories. The probability distribution then can be utilized as the model's confidence indicator. For example, if the confidence model is a classifier having two classes, A, and ~A, wherein the class A can be for instance that step 2 processing will be delivering high confidence results. This kind of classifier can be trained when step 2 ground truth is available at training time., for example as disclosed in Fig. 14.

[0146] Fig. 14 discloses a training arrangement for the model 1310, or at least for the confidence model 1312 in implementations in which the models are trained separately. Referring to Fig. 14, labeled data 1401, e.g. data representing measurement results, is input to the model 1310 (to the model 811’ and to the confidence model 1312). Further, since a ground truth 1404 to step 2 model 1112 output 1403 (the model 1310 output) is available at the training time the ground truth 1404 is input to the confidence model 1312 to train the confidence model to output a confidence estimate of the step 2 model output. In other words, the ground truth for the output of the step 2 model is used in the training of the confidence model 1312. The thus obtained confidence model is trained to estimate confidence of predictions 903.

[0147] In another implementation, the confidence model is not part of the step 1 model, as illustrated in Fig. 13, but a separate model.

[0148] The model 1310 in Fig. 13 may be adjusted to implement the example of Fig. 6 or Fig. 7 by using the quality assessment information 901 as described with Fig. 6 or Fig.7, and to input when the step 2 model is to be performed intermediate results 802 to the step 2 model 1112 and the quality assessment information 901 to the parameter selection portion 1113. However, the adjustment is not illustrated in Fig. 13.

[0149] The above described step 1 models and step 2 models or a model comprising them may be trained in a multiple ways and for different purposes. For example, A3 events may be used as ground truth with corresponding labeled data representing measurement results. The models may be trained to be coverage area -specific, for example for a cell edge, using data collected on mobile apparatuses on the cell edge. Such models may be retrained for a different coverage area. Further, it may be that the step 1 model is trained first, and then the step 2 model is trained, or they are trained together. The training and / or retraining may be performed offline at a mobile apparatus, or online at the mobile apparatus, or the wireless network side may perform the training and provide the mobile apparatus with the trained models.

[0150] The blocks, related functions, and information exchange (messages, signaling) described above by means of Fig. 2 to Fig. 14 are in no absolute chronological order, and some of them may be performed simultaneously or in an order differing from the given one. Other functions can also be executed between them or within them, and other information may be sent, and / or other rules applied. Some of the blocks or part of the blocks or one or more pieces of information can also be left out or replaced by a corresponding block or part of the block or one or more pieces of information. Further, some of the processes, for example the second process performed in the mobile apparatus, may be performed in the network apparatus.

[0151] Fig. 15 illustrates an apparatus 1501 that may be configured to perform at least the first process relating to multi-step process according to some embodiments. The apparatus 1501 may be a mobile apparatus, examples of which are listed above. Fig. 16 illustrates an apparatus that may implement distributed functionality of the apparatus illustrated in Fig. 15. Fig. 17 illustrates an apparatus, e.g. a network apparatus, that may be configured to at least transmit to mobile apparatuses configurations relating to multi-step process. Fig. 18 illustrates an apparatus that may implement distributed functionality of the apparatus illustrated in Fig. 17.

[0152] The apparatus 1501, 1701 may comprise one or more communication control circuitry 1520, 1720, such as at least one processor, and at least one memory 1530, including one or more algorithms 1531, 1731, such as a computer program code (software) wherein the at least one memory and the computer program code (software) are configured, with the at least one processor, to cause the apparatus to carry out any one of the exemplified functionalities of a corresponding apparatus, described above with any of Fig. 1 to Fig. 14. Said at least one memory 1530, 1730 may also comprise at least one database 1532, 1732.

[0153] Referring to Fig. 15, the one or more communication control circuitries 1520 of the apparatus 1501 comprise at least a multi-step circuitry 1521 which is configured to perform multi-step process related functionalities, e.g. perform the first process and adjust the multi-step process according to the configuration. To this end, the multi-step circuitry 1521 of the apparatus 1501 is configured to carry out at least some of the functionalities described above, e.g., by means of Fig. 2 and Fig. 4 to Fig. 14, using one or more individual circuitries.

[0154] Referring to Fig. 15, the memory 1530 may be implemented using any suitable data storage technology, such as semiconductor based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory.

[0155] Referring to Fig. 15, the apparatus 1501 may further comprise different interfaces 1510 such as one or more communication interfaces (TX / RX) comprising hardware and / or software for realizing communication connectivity according to one or more communication protocols. The one or more communication interfaces 1510 may enable connecting to the Internet and / or to a core network of a wireless communications network and / or to a radio access network and / or to other apparatuses within range of the apparatus. The one or more communication interface 1510 may provide the apparatus with communication capabilities to communicate in a cellular communication system and enable communication to different network nodes or elements. The one or more communication interfaces 1510 may comprise standard well-known components such as an amplifier, filter, frequency-converter, (de)modulator, and encoder / decoder circuitries, controlled by the corresponding controlling units, and possibly one or more antennas.

[0156] In an embodiment, as shown in Fig. 16, at least some of the functionalities of the apparatus of Fig. 15 may be shared between two physically separate devices, forming one operational entity. Therefore, the apparatus may be seen to depict the operational entity comprising one or more physically separate devices for executing at least some of the described processes. Thus, the apparatus of Fig. 16, utilizing such shared architecture, may comprise a control unit CU 1620, or a remote control unit, such as a host computer or a server computer, operatively coupled (e.g. via a wireless or wired connection) to a remote distributed unit DU 1622 located in the mobile apparatus or in a remote head of the mobile apparatus. In an embodiment, at least some of the described processes may be performed by the CU 1620. In an embodiment, the execution of at least some of the described processes may be shared among the DU 1622 and the CU 1620.

[0157] Similar to Fig. 15, the apparatus of Fig. 16 may comprise one or more communication control circuitries (CNTL) 1520, such as at least one processor, and at least one memory (MEM) 1530, including one or more algorithms (PROG) 1531, such as a computer program code (software) wherein the at least one memory and the computer program code (software) are configured, with the at least one processor, to cause the apparatus to carry out any one of the exemplified functionalities of the mobile apparatus described above, e.g., by means of Fig. 1, Fig. 2 and Fig. 4 to Fig. 14.

[0158] Referring to Fig. 17, the one or more communication control circuitry 1720 of the apparatus 1701 comprise at least a configuring circuitry 1721, which is configured to include into its transmissions configurations, that are discussed with Fig. 2 to Fig. 14. To this end, the configuring circuitry 1721 of the apparatus 1701 is configured to carry out at least some of the functionalities of the access network apparatus, described above, e.g., by means of Fig. 1, and Fig. 3 to Fig.14, for example, by the apparatus transmitting configurations, using one or more individual circuitries.

[0159] Referring to Fig. 17, the memory 1730 may be implemented using any suitable data storage technology, such as semiconductor based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory.

[0160] Referring to Fig. 17, the apparatus 1701 may further comprise different interfaces 1710 such as one or more communication interfaces (TX / RX) comprising hardware and / or software for realizing communication connectivity according to one or more communication protocols. The one or more communication interfaces 1710 may enable connecting to the Internet and / or to a core network of a wireless communications network. The one or more communication interface 1710 may provide the apparatus with communication capabilities to communicate in a cellular communication system and enable communication to different network nodes or elements or mobile apparatuses, e.g., terminal devices or user equipments, for example. The one or more communication interfaces 1710 may comprise standard well-known components such as an amplifier, filter, frequency-converter, (de)modulator, and encoder / decoder circuitries, controlled by the corresponding controlling units, and one or more antennas.

[0161] In an embodiment, as shown in Fig. 18, at least some of the functionalities of the apparatus of Fig. 17 may be shared between two physically separate devices, forming one operational entity. Therefore, the apparatus may be seen to depict the operational entity comprising one or more physically separate devices for executing at least some of the described processes. Thus, the apparatus of Fig. 18, utilizing such shared architecture, may comprise a control unit CU 1820, or a remote control unit, such as a host computer or a server computer, operatively coupled (e.g. via a wireless or wired network) to a remote distributed unit DU 1822 located in a access network apparatus, e.g., a base station. In an embodiment, at least some of the described processes may be performed by the CU 1820. In an embodiment, the execution of at least some of the described processes may be shared among the DU 1822 and the CU 1820.

[0162] Similar to Fig. 17, the apparatus of Fig. 18 may comprise one or more communication control circuitries (CNTL) 1720, such as at least one processor, and at least one memory (MEM) 1730, including one or more algorithms (PROG) 1731, such as a computer program code (software) wherein the at least one memory and the computer program code (software) are configured, with the at least one processor, to cause the apparatus to carry out any one of the exemplified functionalities of the apparatus described above, e.g., by means of Fig. 1, and Fig. 3 to Fig. 14, for example, by the access network apparatus.

[0163] In embodiments, the CU 1620, 1820 may generate a virtual network through which the CU 1620, 1820 communicates with the DU 1622, 1822. In general, virtual networking may involve a process of combining hardware and software network resources and network functionality into a single, softwarebased administrative entity, a virtual network. Network virtualization may involve platform virtualization, often combined with resource virtualization. Network virtualization may be categorized as external virtual networking which combines many networks, or parts of networks, into the server computer or the host computer (e.g. to the CU). External network virtualization is targeted to optimized network sharing. Another category is internal virtual networking which provides network-like functionality to the software containers on a single system. Virtual networking may also be used for testing the terminal device.

[0164] In embodiments, the virtual network may provide flexible distribution of operations between the DU and the CU. In practice, any digital signal processing task may be performed in either the DU or the CU and the boundary where the responsibility is shifted between the DU and the CU may be selected according to implementation.

[0165] As used in this application, the term ‘circuitry’ may refer to one or more or all of the following: (a) hardware-only circuit implementations, such as implementations in only analog and / or digital circuitry, and (b) combinations of hardware circuits and software (and / or firmware), such as (as applicable): (i) a combination of analog and / or digital hardware circuit(s) with software / firmware and (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 terminal device or an access node, to perform various functions, and (c) hardware circuit(s) and 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 any claims. As a further example, as used in this application, the term ‘circuitry’ also covers an implementation of merely a hardware circuit or processor (or multiple processors) or a 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 for an access node or a terminal device or other computing or network device.

[0166] In an embodiment, at least some of the processes described in connection with Fig. 2 to Fig. 14 may be carried out by an apparatus comprising corresponding means for carrying out at least some of the described processes. Some example means for carrying out the processes may include at least one of the following: detector, processor (including dual-core and multiple-core processors), digital signal processor, controller, receiver, transmitter, encoder, decoder, memory, RAM, ROM, software, firmware, display, user interface, display circuitry, user interface circuitry, user interface software, display software, circuit, antenna, antenna circuitry, and circuitry. In an embodiment, the at least one processor, the memory, and the computer program code form processing means or comprises one or more computer program code portions for carrying out one or more operations according to any one of the embodiments of Fig. 2 to Fig. 14 or operations thereof.

[0167] Embodiments and examples as described may also be carried out in the form of a computer process defined by a computer program or portions thereof. Embodiments of the functionalities described in connection with Fig. 2 to Fig. 14 may be carried out by executing at least one portion of a computer program comprising corresponding instructions. The computer program may be provided as a computer readable medium comprising program instructions stored thereon or as a non-transitory computer readable medium comprising program instructions stored thereon. The computer program may be in source code form, object code form, or in some intermediate form, and it may be stored in some sort of carrier, which may be any entity or device capable of carrying the program. For example, the computer program may be stored on a computer program distribution medium readable by a computer or a processor. The computer program medium may be, for example but not limited to, a record medium, computer memory, read-only memory, electrical carrier signal, telecommunications signal, and software distribution package, for example. The computer program medium may be a non-transitory medium. The term “non- transitory,” as used herein, is a limitation of the medium itself (i.e., tangible, not a signal ) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM). Coding of software for carrying out the embodiments as shown and described is well within the scope of a person of ordinary skill in the art.

[0168] Even though the embodiments have been described above with reference to examples according to the accompanying drawings, it is clear that the embodiments are not restricted thereto but can be modified in several ways within the scope of the appended claims. Therefore, all words and expressions should be interpreted broadly and they are intended to illustrate, not to restrict, the embodiment. It will be obvious to a person skilled in the art that, as technology advances, the inventive concept can be implemented in various ways. Further, it is clear to a person skilled in the art that the described embodiments may, but are not required to, be combined with other embodiments in various ways.

Claims

CLAIMS1. An apparatus comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: receive from a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; obtain measurement results of reference signals received by the apparatus; provide the measurement results as input to the first process; and adjust, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

2. The apparatus of claim 1, wherein the adjusting the multi-step process comprises: modifying the intermediate results output by the first process; and inputting modified intermediate results to the second process.

3. The apparatus of claim 1 or 2, wherein the adjusting the multi-step process comprises: inputting to the second process also at least part of the quality assessment information or values determined from the quality assessment information.

4. The apparatus of claim 1, 2 or 3, wherein the adjusting the multi-step process comprises: transmitting to the wireless network at least part of the quality assessment information; receiving from the wireless network an indication indicating whether or not to continue the multi-step process; and based on the indication indicating to continue, perform, according to the configuration, the second process.

5. The apparatus of claim 1, 2 or 3, wherein the adjusting the multi-step process comprises: determining whether to perform, according to the configuration, the second process, or to transmit to the wireless network the measurement results provided as input to the first process; and based on a result of the determining, either perform the second process, or transmit to the wireless network the measurement results.

6. The apparatus of claim 1, 2 or 3, wherein the adjusting the multi-step process comprises: determining whether to or not perform, according to the configuration, the second process; and based on a result of the determining, either perform or not to perform the second process.

7. The apparatus of any preceding claim, wherein the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive in the configuration an indication on quality assessment method to be used; and select a first process based on said indication.

8. The apparatus of any preceding claim, wherein the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: transmit to the wireless network a request for the configuration; and receive the configuration as a response to the request.

9. The apparatus of any preceding claim, wherein the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: transmit to the wireless network information at least on the predictions.

10. The apparatus of any preceding claim, wherein the predictions are mobility predictions.

11. The apparatus of any preceding claim, wherein the configuration indicates whether to use, for the quality assessment information, accuracy of intermediate results of the first process, and / or confidence estimate of intermediate results of the first process and / or confidence estimate of the predictions of the second process.

12. An apparatus comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: transmit over a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus.

13. The apparatus of claim 12, wherein the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive from a second apparatus of the at least one second apparatus at least part of the quality assessment information; determine, based on the at least part of quality assessment information, whether or not the second apparatus is to continue the multi-step process; and transmit to the second apparatus an indication indicating whether or not to continue the multi-step process.

14. The apparatus of claim 12 or 13, wherein the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: determine, prior to transmitting, the configuration.

15. The apparatus of claim 14, wherein the at least one processor and the at least one memory storing instructions, when executed by the at least one processor, further cause the apparatus at least to: receive from at least one of the at least one second apparatus a request for the configuration; and perform the determining and transmitting the configuration in response to the request being received.

16. A method comprising at least: receiving, by an apparatus, from a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; obtaining measurement results of reference signals received by the apparatus; providing the measurement results as input to the first process; and adjusting, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

17. The method of claim 16, wherein the adjusting the multi-step process comprises: modifying the intermediate results output by the first process; and inputting modified intermediate results to the second process.

18. The method of claim 16 or 17, wherein the adjusting the multi-step process comprises: inputting to the second process also at least part of the quality assessment information or values determined from the quality assessment information.

19. The method of claim 16, 17 or 18, wherein the adjusting the multi- step process comprises:transmitting to the wireless network at least part of the quality assessment information; receiving from the wireless network an indication indicating whether or not to continue the multi-step process; and based on the indication indicating to continue, perform, according to the configuration, the second process.

20. The method of claim 16, 17 or 18, wherein the adjusting the multi- step process comprises: determining whether to perform, according to the configuration, the second process, or to transmit to the wireless network the measurement results provided as input to the first process; and based on a result of the determining, either perform the second process, or transmit to the wireless network the measurement results.

21. The method of claim 16, 17 or 18, wherein the adjusting the multi- step process comprises: determining whether to or not perform, according to the configuration, the second process; and based on a result of the determining, either perform or not to perform the second process.

22. The method of any of claims 16 to 21, further comprising at least: receiving in the configuration an indication on quality assessment method to be used; and selecting a first process based on said indication.

23. The method of any of claims 16 to 22, further comprising at least: transmitting to the wireless network a request for the configuration; and receiving the configuration as a response to the request.

24. The method of any of claims 16 to 23, further comprising at least: transmitting to the wireless network information at least on the predictions.

25. The method of any of claims 16 to 24, wherein the predictions are mobility predictions.

26. The method of any of claims 16 to 25, wherein the configuration indicates whether to use, for the quality assessment information, accuracy of intermediate results of the first process, and / or confidence estimate of intermediate results of the first process and / or confidence estimate of the predictions of the second process.

27. A method comprising at least: transmitting over a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus.

28. The method of claim 27, further comprising at least: receiving from a second apparatus of the at least one second apparatus at least part of the quality assessment information; determining, based on the at least part of quality assessment information, whether or not the second apparatus is to continue the multi-step process; and transmitting to the second apparatus an indication indicating whether or not to continue the multi-step process.

29. The method of claim 27 or 28, further comprising: determining, prior to transmitting, the configuration.

30. The method of claim 29, further comprising: receiving from at least one of the at least one second apparatus a request for the configuration; and performing the determining and transmitting the configuration in response to the request being received.

31. A computer readable medium comprising instructions which, when executed by an apparatus, cause the apparatus to perform at least the following: receiving from a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; obtaining measurement results of reference signals received by the apparatus; providing the measurement results as input to the first process; and adjusting, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

32. The computer readable medium of claim 31, further comprising instructions which, when executed by an apparatus, cause the apparatus to perform the method of any of claims 17 to 26.

33. A computer readable medium comprising instructions which, when executed by an apparatus, cause the apparatus to perform at least the following: transmitting over a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus.

34. The computer readable medium of claim 33, further comprising instructions which, when executed by an apparatus, cause the apparatus to perform the method of any of claims 28 to 30.

35. The computer readable medium of claim 31, 32, 33 or 34, wherein the computer readable medium is a non-transitory computer readable medium.

36. A computer program comprising instructions, which, when executed by an apparatus, cause the apparatus to perform the method of any of claims 16 to 30.

37. An apparatus comprising at least: means for receiving from a wireless network a configuration for a multi- step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions; means for obtaining measurement results of reference signals received by the apparatus; means for providing the measurement results as input to the first process; and means for adjusting, according to the configuration, based at least on the quality assessment information of the intermediate results outputted by the first process, the multi-step process.

38. The apparatus of claim 37, wherein the means for adjusting the multi-step process comprises: means for modifying the intermediate results output by the first process; and means for inputting modified intermediate results to the second process.

39. The apparatus of claim 37 or 38, wherein the means for adjusting the multi-step process comprises: means for inputting to the second process also at least part of the quality assessment information or values determined from the quality assessment information.

40. The apparatus of claim 37, 38 or 39, wherein the means for adjusting the multi-step process comprises: means for transmitting to the wireless network at least part of the quality assessment information;means for receiving from the wireless network an indication indicating whether or not to continue the multi-step process; and means for performing, based on the indication indicating to continue, the second process according to the configuration.

41. The apparatus of claim 37, 38 or 39, wherein the means for adjusting the multi-step process comprises: means for determining whether to perform, according to the configuration, the second process, or to transmit to the wireless network the measurement results provided as input to the first process; means for performing the second process, when a result of the means for determining is to perform the second process; and means for transmitting to the wireless network the measurement results, when a result of the means for determining is not to perform the second process.

42. The apparatus of claim 37, 38 or 39, wherein the means for adjusting the multi-step process comprises: means for determining whether to or not perform, according to the configuration, the second process; means for performing the second process, when a result of the means for determining is to perform the second process; and means for not performing the second process, when a result of the means for determining is not to perform the second process.

43. The apparatus of any of claims 37 to 42, further comprising: means for receiving in the configuration an indication on quality assessment method to be used; and means for selecting a first process based on said indication.

44. The apparatus of any of claims 37 to 43, further comprising: means for transmitting to the wireless network a request for the configuration; and means for receiving the configuration as a response to the request.

45. The apparatus of any of claims 37 to 44, further comprising: means for transmitting to the wireless network information at least on the predictions.

46. The apparatus of any of claims 37 to 45, wherein the predictions are mobility predictions.

47. The apparatus of any of claims 37 to 46, wherein the configuration indicates whether to use, for the quality assessment information, accuracy of intermediate results of the first process, and / or confidence estimate of intermediate results of the first process and / or confidence estimate of the predictions of the second process.

48. An apparatus comprising at least: means for transmitting over a wireless network a configuration for a multi-step process, which comprises at least a first process outputting at least intermediate results for at least one channel characteristics and quality assessment information of the intermediate results, and a second process, which is a subsequent process to the first process using at least inputs that are based on the intermediate results and outputting predictions to at least one second apparatus.

49. The apparatus of claim 48, further comprising at least: means for receiving from a second apparatus of the at least one second apparatus at least part of the quality assessment information; means for determining, based on the at least part of quality assessment information, whether or not the second apparatus is to continue the multi-step process; and means for transmitting to the second apparatus an indication indicating whether or not to continue the multi-step process.

50. The apparatus of claim 48 or 49, further comprising: means for determining, prior to transmitting, the configuration.

51. The apparatus of claim 50, further comprising: means for receiving from at least one of the at least one second apparatus a request for the configuration;wherein the means for determining and the means for transmitting are responsive to the request being received.

Images