Downlink reference signal configuration

Abstract

Disclosed is a method performed by a user equipment, the method comprising: receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitoring the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring.

Description

[0001]DOWNLINK^REFERENCE^SIGNAL^CONFIGURATION FIELD The following example embodiments relate to wireless communication. BACKGROUND Channel estimation is a process that involves determining the characteristicsof a wireless communication channel, such as the path loss, fading, and / or interference.The channel estimate information may be used by the receiver to accurately decode the received data signal. SUMMARY The scope of protection sought for various example embodiments is set out by the claims. The example embodiments and features, if any, described in this specification that do not fall under the scope of the claims are to be interpreted as examples useful for understanding various embodiments. According to a first aspect, there is provided a user equipment comprising:means for generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configurationused for training of a channel predictor associated with the channel estimation; meansfor transmitting the capability indication to a network node; and means for receiving,from the network node, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal inat least one of a frequency domain or a time domain, the transmission patterncomprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal. According to a second aspect, there is provided the user equipment of the first aspect, wherein the transmission pattern further comprises a second set of resource elements on which the network node does not intend to transmit the downlink reference signal, wherein the first set of resource elements is associated with one or more first antenna ports that are equivalent to or different than one or more second antenna ports associated with the second set of resource elements. According to a third aspect, there is provided the user equipment of thesecond aspect, further comprising: means for receiving, from the network node, thedownlink reference signal on the first set of resource elements; means for determining, per antenna port associated with the downlink reference signal, one or more channel estimates associated with the first set of resource elements in at least one of the frequency domain or the time domain; and means for determining, using the channel predictor, per antenna port associated with the downlink reference signal, one or more predicted channel estimates associated with the second set of resource elements in at least one of the frequency domain or the time domain. According to a fourth aspect, there is provided the user equipment of the third aspect, further comprising means for demodulating a downlink data channel associated with the downlink reference signal based on a combination of the one or more channel estimates and the one or more predicted channel estimates. According to a fifth aspect, there is provided the user equipment of the third or fourth aspect, wherein the downlink reference signal is not received on the second set of resource elements in at least one of the frequency domain or the time domain. According to a sixth aspect, there is provided the user equipment of any of thesecond to fifth aspects, further comprising means for receiving a downlink signal fromthe network node on at least one resource element of the second set of resource elements in at least one of the frequency domain or the time domain, wherein the downlink signal comprises at least one of: a data signal, a control signal, or another reference signal different from the downlink reference signal for which the downlink reference signal configuration is received. According to a seventh aspect, there is provided the user equipment of any of the first to sixth aspects, wherein the channel predictor comprises a machine learning model pre-trained based on a set of input data and a set of expected output data, wherein the set of expected output data comprises one or more reference signal channel estimates in at least one of the frequency domain or the time domain per antenna port associated with the downlink reference signal, wherein the set of input data comprises at least one of: a physical resource block level granularity associated with a resource element pattern type of the downlink reference signal in the frequency domain, a symbol position pattern of the downlink reference signal in the time domain, a number of antenna ports associated with the downlink reference signal, a sequence type of the downlink reference signal with one or more initialization seed values, or a downlink precoding technique of the downlink reference signal. According to an eighth aspect, there is provided the user equipment of any of the first to seventh aspects, wherein the downlink reference signal configuration further comprises at least one of: an indication for operating the channel predictor in at least one of the frequency domain or the time domain, or at least one data set identifier from the one or more data set identifiers to be applied for the channel predictor. According to a ninth aspect, there is provided the user equipment of any of the first to eighth aspects, wherein the one or more data set identifiers indicate at least one physical resource block density allocation of a reference signal resource element pattern type used for training the channel predictor. According to a tenth aspect, there is provided the user equipment of any ofthe first to eighth aspects, wherein the one or more data set identifiers indicate at leastone physical resource block density allocation of a reference signal resource element pattern type that the channel predictor is capable of predicting. According to an eleventh aspect, there is provided the user equipment of any of the first to tenth aspects, wherein the one or more data set identifiers indicate at least one reference signal symbol position pattern that the channel predictor is capable of predicting in the time domain. According to a twelfth aspect, there is provided the user equipment of any of the first to eleventh aspects, wherein the one or more data set identifiers indicate at least one of: one or more carrier frequencies supported by the channel predictor, one or more numerology options supported by the channel predictor, one or more delay spread ranges supported by the channel predictor, one or more Doppler frequency shift or spread values supported by the channel predictor, one or more user equipment speed values supported by the channel predictor, one or more reference signal sequence initialization seed values supported by the channel predictor, a number of reference signal antenna ports supported by the channel predictor, one or more reference signal resource element types supported by the channel predictor, one or more reference signal sequence types supported by the channel predictor, a physical downlink shared channel allocation length or range, in at least one of the frequency domain or the time domain, supported by the channel predictor, one or more precoding types supported by the channel predictor, or a precoding granularity, in at least one of the frequency domain or the time domain, supported by the channel predictor. According to a thirteenth aspect, there is provided the user equipment of any of the first to twelfth aspects, wherein the downlink reference signal comprises one of: a physical downlink shared channel demodulation reference signal, a channel state information reference signal, or a phase-tracking reference signal. According to a fourteenth aspect, there is provided a network node comprising: means for receiving, from a user equipment, a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; means for determining, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal; and means for transmitting the downlink reference signal configuration to the user equipment. According to a fifteenth aspect, there is provided the network node of the fourteenth aspect, wherein the transmission pattern further comprises a second set of resource elements on which the network node does not intend to transmit the downlink reference signal. According to a sixteenth aspect, there is provided the network node of the fifteenth aspect, further comprising: means for transmitting the downlink reference signal to the user equipment on the first set of resource elements; and means for transmitting a downlink signal to the user equipment on at least one resource element of the second set of resource elements in at least one of the frequency domain or the timedomain, wherein the downlink signal comprises at least one of: a data signal, a controlsignal, or another reference signal different from the downlink reference signal for whichthe downlink reference signal configuration is transmitted. According to a seventeenth aspect, there is provided a method performed by a user equipment, the method comprising: generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; transmitting the capability indication to a network node; and receiving, from the network node, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal. According to an eighteenth aspect, there is provided a method performed by a network node, the method comprising: receiving, from a user equipment, a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; determining, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal; and transmitting the downlink reference signal configuration to the user equipment. According to a nineteenth aspect, there is provided a computer program comprising instructions which, when executed by a user equipment, cause the user equipment to perform at least the following: generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; transmitting the capability indication to a network node; and receiving, from the network node, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal. According to a twentieth aspect, there is provided a computer program comprising instructions which, when executed by a network node, cause the networknode to perform at least the following: receiving, from a user equipment, a capabilityindication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more dataset identifiers indicating a reference signal configuration used for training of a channelpredictor associated with the channel estimation; determining, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one of a frequencydomain or a time domain, the transmission pattern comprising at least a first set ofresource elements on which the network node intends to transmit the downlink reference signal; and transmitting the downlink reference signal configuration to the user equipment. According to a twenty-first aspect, there is provided a non-transitorycomputer readable medium comprising program instructions which, when executed by a user equipment, cause the user equipment to perform at least the following: generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; transmitting the capability indication to a network node; and receiving, from the network node, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal. According to a twenty-second aspect, there is provided a non-transitory computer readable medium comprising program instructions which, when executed bya network node, cause the network node to perform at least the following: receiving, from a user equipment, a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; determining, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal; and transmitting the downlink reference signal configuration to the user equipment. According to a twenty-third aspect, there is provided a computer readable medium comprising program instructions which, when executed by a user equipment, cause the user equipment to perform at least the following: generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; transmitting the capability indication to a network node; and receiving, from the network node, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intends to transmit the downlink reference signal. According to a twenty-fourth aspect, there is provided a computer readablemedium comprising program instructions which, when executed by a network node, cause the network node to perform at least the following: receiving, from a user equipment, a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; determining, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising at least a first set of resource elements on which the network node intendsto transmit the downlink reference signal; and transmitting the downlink referencesignal configuration to the user equipment. According to a twenty-fifth aspect, there is provided a user equipment comprising: means for receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; means for monitoring the second set of resource elementsfor a downlink data channel associated with the downlink reference signal; and meansfor receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring. According to a twenty-sixth aspect, there is provided the user equipment of the twenty-fifth aspect, wherein the means for the monitoring are configured to performthe monitoring based on the user equipment being configured to perform channelprediction associated with channel estimation on the second set of resource elements. According to a twenty-seventh aspect, there is provided the user equipment of the twenty-fifth or twenty-sixth aspect, further comprising means for generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; and means for transmitting the capability indication to the network node, wherein the downlink reference signal configuration is based on the capability indication. According to a twenty-eight aspect, there is provided the user equipment of the twenty-seventh aspect, further comprising: means for receiving, from the network node, the downlink reference signal on the first set of resource elements; means for determining, per antenna port associated with the downlink reference signal, one or more channel estimates associated with the first set of resource elements in at least one of the frequency domain or the time domain; means for determining, using the channel predictor, per antenna port associated with the downlink reference signal, one or more predicted channel estimates associated with the second set of resource elements in at least one of the frequency domain or the time domain; and means for demodulating the downlink data channel based on a combination of the one or more channel estimates and the one or more predicted channel estimates. According to a twenty-ninth aspect, there is provided the user equipment of the twenty-seventh or twenty-eighth aspect, wherein the channel predictor comprises a machine learning model pre-trained based on a set of input data and a set of expected output data, wherein the set of expected output data comprises one or more reference signal channel estimates in at least one of the frequency domain or the time domain per antenna port associated with the downlink reference signal, wherein the set of input data comprises at least one of: a physical resource block level granularity associated with a resource element pattern type of the downlink reference signal in the frequency domain, a symbol position pattern of the downlink reference signal in the time domain, a number of antenna ports associated with the downlink reference signal, a sequence type of the downlink reference signal with one or more initialization seed values, or a downlink precoding technique of the downlink reference signal. According to a thirtieth aspect, there is provided the user equipment of any of the twenty-seventh to twenty-ninth aspects, wherein the downlink reference signal configuration further comprises at least one of: an indication for operating the channel predictor in at least one of the frequency domain or the time domain, or at least one data set identifier from the one or more data set identifiers to be applied for the channel predictor. According to a thirty-first aspect, there is provided the user equipment of anyof the twenty-seventh to thirtieth aspects, wherein the one or more data set identifiersindicate at least one physical resource block density allocation of a reference signal resource element pattern type used for training the channel predictor. According to a thirty-second aspect, there is provided the user equipment ofany of the twenty-seventh to thirtieth aspects, wherein the one or more data setidentifiers indicate at least one physical resource block density allocation of a reference signal resource element pattern type that the channel predictor is capable of predicting. According to a thirty-third aspect, there is provided the user equipment of any of the twenty-seventh to thirty-second aspects, wherein the one or more data set identifiers indicate at least one of: one or more carrier frequencies supported by the channel predictor, one or more numerology options supported by the channel predictor, one or more delay spread ranges supported by the channel predictor, one or more Doppler frequency shift or spread values supported by the channel predictor, one or more user equipment speed values supported by the channel predictor, one or more reference signal sequence initialization seed values supported by the channel predictor, a number of reference signal antenna ports supported by the channel predictor, one or more reference signal resource element types supported by the channel predictor, one or more reference signal sequence types supported by the channel predictor, a physical downlink shared channel allocation length or range, in at least one of the frequency domain or the time domain, supported by the channel predictor, one or more precoding types supported by the channel predictor, or a precoding granularity, in at least one of the frequency domain or the time domain, supported by the channel predictor. According to a thirty-fourth aspect, there is provided the user equipment of any of the twenty-fifth to thirty-third aspects, wherein the downlink reference signal is a physical downlink shared channel demodulation reference signal, wherein the downlink data channel is a physical downlink shared channel. According to a thirty-fifth aspect, there is provided a network node comprising: means for generating a downlink reference signal configuration, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; means for transmitting the downlink reference signal configuration to a user equipment; and means for transmitting, to the user equipment, on the second set of resource elements, a downlink data channel associated with the downlink reference signal. According to a thirty-sixth aspect, there is provided a method performed by a user equipment, the method comprising: receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitoring the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring. According to a thirty-seventh aspect, there is provided the method of the thirty-sixth aspect, wherein the monitoring is performed based on the user equipment being configured to perform channel prediction associated with channel estimation on the second set of resource elements. According to a thirty-eighth aspect, there is provided the method of the thirty- sixth or thirty-seventh aspect, further comprising: generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictorassociated with the channel estimation; and transmitting the capability indication to thenetwork node, wherein the downlink reference signal configuration is based on the capability indication. According to a thirty-ninth aspect, there is provided a method performed bya network node, the method comprising: generating a downlink reference signalconfiguration, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; transmitting the downlink reference signal configuration to a user equipment; and transmitting, to the user equipment, on the second set of resource elements, a downlink data channel associated with the downlink reference signal. According to a fortieth aspect, there is provided a computer programcomprising instructions which, when executed by a user equipment, cause the user equipment to perform at least the following: receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitoring the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring. According to a forty-first aspect, there is provided a computer program comprising instructions which, when executed by a network node, cause the networknode to perform at least the following: generating a downlink reference signalconfiguration, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; transmitting the downlink reference signal configuration to a user equipment; and transmitting, to the user equipment, on the second set of resource elements, a downlink data channel associated with the downlink reference signal. According to a forty-second aspect, there is provided a non-transitorycomputer readable medium comprising program instructions which, when executed by a user equipment, cause the user equipment to perform at least the following: receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitoring the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring. According to a forty-third aspect, there is provided a non-transitory computer readable medium comprising program instructions which, when executed by a network node, cause the network node to perform at least the following: generating a downlink reference signal configuration, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; transmitting the downlink reference signal configuration to a user equipment; and transmitting, to the user equipment, on the second set of resource elements, a downlink data channel associated with the downlink reference signal. According to a forty-fourth aspect, there is provided a computer readable medium comprising program instructions which, when executed by a user equipment, cause the user equipment to perform at least the following: receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitoring the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring. According to a forty-fifth aspect, there is provided a computer readable medium comprising program instructions which, when executed by a network node, cause the network node to perform at least the following: generating a downlink reference signal configuration, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; transmitting the downlink reference signal configuration to a user equipment; and transmitting, to the user equipment, on the second set of resource elements, a downlink data channel associated with the downlink reference signal.BRIEF DESCRIPTION OF THE DRAWINGSIn the following, various example embodiments will be described in greater detail with reference to the accompanying drawings, in which FIG. 1 illustrates an example of a wireless communication network;FIG.2A illustrates the training phase of a machine learning model; FIG.2B illustrates the inference phase of a trained machine learning model; FIG.3 illustrates a signal flow diagram; FIG. 4 illustrates three examples of downlink reference signal resourceelement patterns; FIG.5 illustrates a flow chart; FIG.6 illustrates a flow chart; FIG.7 illustrates a flow chart; FIG.8 illustrates a flow chart; FIG. 9 illustrates an example of an apparatus; andFIG.10 illustrates an example of an apparatus. DETAILED DESCRIPTION The following embodiments are exemplifying. Although the specification may refer to “an”, “one”, or “some” embodiment(s) in several locations of the text, this does not necessarily mean that each reference is made to the same embodiment(s), or that a particular feature only applies to a single embodiment. Single features of differentembodiments may also be combined to provide other embodiments within the scope ofthe claims. Furthermore, the words "comprising" and "including" should be understoodas not limiting the described embodiments to consist of only those features that have been mentioned, and such embodiments may also contain features that have not beenspecifically mentioned. Reference numbers, in the description and / or in the claims, serveto illustrate the embodiments with reference to the drawings, without limiting theembodiments to these examples only. Some example embodiments described herein may be implemented in a wireless communication network comprising a radio access network based on one ormore of the following radio access technologies (RATs): global system for mobilecommunications (GSM) or any other second generation (2G) radio access technology,universal mobile telecommunication system (UMTS, 3G) based on basic wideband-code division multiple access (W-CDMA), high-speed packet access (HSPA), long termevolution (LTE), LTE-Advanced, fourth generation (4G), fifth generation (5G), 5G newradio (NR), 5G-Advanced (i.e., 3GPP NR Rel-18 and beyond), or sixth generation (6G). Some examples of radio access networks include the universal mobile telecommunications system (UMTS) radio access network (UTRAN), the evolved universal terrestrial radio access network (E-UTRA), or the next generation radio accessnetwork (NG-RAN). The wireless communication network may further comprise a corenetwork, and some example embodiments may also be applied to network functions ofthe core network. It should be noted that the embodiments are not restricted to the wirelesscommunication network given as an example, but a person skilled in the art may alsoapply the solution to other wireless communication networks or systems provided withnecessary properties. For example, some example embodiments may also be applied toa communication system based on IEEE 802.11 specifications, or a communicationsystem based on IEEE 802.15 specifications. IEEE is an abbreviation for the Institute ofElectrical and Electronics Engineers. FIG. 1 depicts an example of a simplified wireless communication network showing some physical and logical entities. The connections shown in FIG. 1 may be physical connections or logical connections. It is apparent to a person skilled in the art that the wireless communication network may also comprise other physical and logical entities than those shown in FIG.1. The example embodiments described herein are not, however, restricted to the wireless communication network given as an example but a person skilled in the art may apply the example embodiments described herein to other wireless communication networks provided with necessary properties. The example wireless communication network shown in FIG. 1 includes a radio access network (RAN) and a core network 110. FIG. 1 shows user equipment (UE) 100, 102 configured to be in a wirelessconnection on one or more communication channels in a radio cell with an access node104 of a radio access network.The access node 104 may comprise a computing device configured to controlthe radio resources of the access node 104 and to be in a wireless connection with oneor more UEs 100, 102. The access node 104 may also be referred to as a base station, abase transceiver station (BTS), an access point, a cell site, a network node, a radio accessnetwork node, a RAN node, or a network device. In this description, the terms “accessnode” and “network node” may be used interchangeably. The access node 104 may be, for example, an evolved NodeB (abbreviated aseNB or eNodeB), or a next generation evolved NodeB (abbreviated as ng-eNB), or a next generation NodeB (abbreviated as gNB or gNodeB), providing the radio cell. The accessnode 104 may include or be coupled to transceivers. From the transceivers of the accessnode 104, a connection may be provided to an antenna unit that establishes a bi-directional radio link to one or more UEs 100, 102. The antenna unit may comprise anantenna or antenna element, or a plurality of antennas or antenna elements. The wireless connection (e.g., radio link) from a UE 100, 102 to the accessnode 104 may be called uplink (UL) or reverse link, and the wireless connection (e.g.,radio link) from the access node 104 to the UE 100, 102 may be called downlink (DL) orforward link. A UE 100 may also communicate directly with another UE 102, and viceversa, via a wireless connection generally referred to as a sidelink (SL). It should beappreciated that the access node 104 or its functionalities may be implemented by usingany node, host, server, access point or other entity suitable for providing such functionalities. The radio access network may comprise more than one access node 104, inwhich case the access nodes may also be configured to communicate with one anotherover wired or wireless links. These links between access nodes may be used for sendingand / or receiving control plane signaling and also for routing data from one access nodeto another access node. The access node 104 may further be connected to a core network (CN) 110. The core network 110 may comprise an evolved packet core (EPC) network and / or a 5thgeneration core network (5GC). The EPC may comprise network entities, such as aserving gateway (S-GW for routing and forwarding data packets), a packet data network gateway (P-GW) for providing connectivity of UEs to external packet data networks,and / or a mobility management entity (MME). The 5GC may comprise one or morenetwork functions, such as at least one of: a user plane function (UPF), an access andmobility management function (AMF), a location management function (LMF), and / or a session management function (SMF). The core network 110 may also be able to communicate with one or more external networks 113, such as a public switched telephone network or the Internet, or utilize services provided by them. For example, in 5G wireless communication networks, the UPF of the core network 110 may be configured to communicate with an external data network via an N6 interface. In LTE wireless communication networks, the P-GW of the core network 110 may be configured to communicate with an external data network. It should also be understood that the distribution of functions between core network operations and access node operations may differ in future wireless communication networks compared to that of the LTE or 5G, or even be non-existent. The illustrated UE 100, 102 is one type of an apparatus to which resources on the air interface may be allocated and assigned. The UE 100, 102 may also be called a wireless communication device, a subscriber unit, a mobile station, a remote terminal,an access terminal, a user terminal, a terminal device, or a user device, just to mentionbut a few names. The UE 100, 102 may be a computing device operating with or withouta subscriber identification module (SIM), including, but not limited to, the following types of computing devices: a mobile phone, a smartphone, a personal digital assistant (PDA), a handset, a computing device comprising a wireless modem (e.g., an alarm or measurement device, etc.), a laptop computer, a desktop computer, a tablet, a game console, a notebook, a multimedia device, a reduced capability (RedCap) device, a wearable device (e.g., a watch, earphones or eyeglasses) with radio parts, a sensor comprising a wireless modem, or a computing device comprising a wireless modem integrated in a vehicle. It should be appreciated that the UE 100, 102 may also be a nearly exclusiveuplink-only device, of which an example may be a camera or video camera loadingimages or video clips to a network. The UE 100, 102 may also be a device havingcapability to operate in an Internet of Things (IoT) network, which is a scenario in whichobjects may be provided with the ability to transfer data over a network without requiring human-to-human or human-to-computer interaction. The wireless communication network may also be able to support the usageof cloud services. For example, at least part of core network operations may be carriedout as a cloud service (this is depicted in FIG.1 by “cloud” 114). The UE 100, 102 mayalso utilize the cloud 114. In some applications, the computation for a given UE may becarried out in the cloud 114 or in another UE.The wireless communication network may also comprise a central control entity, such as a network management system (NMS), or the like. The NMS is a centralized suite of software and hardware used to monitor, control, and administer the network infrastructure. The NMS is responsible for a wide range of tasks such as fault management, configuration management, security management, performance management, and accounting management. The NMS enables network operators to efficiently manage and optimize network resources, ensuring that the network delivers high performance, reliability, and security. 5G enables using multiple-input and multiple-output (MIMO) antennas in theaccess node 104 and / or the UE 100, 102, many more base stations or access nodes than an LTE network (a so-called small cell concept), including macro sites operating in co- operation with smaller stations and employing a variety of radio technologies depending on service needs, use cases and / or spectrum available. 5G wireless communication networks may support a wide range of use cases and related applications including video streaming, augmented reality, different ways of data sharing and various forms ofmachine-type applications, such as (massive) machine-type communications (mMTC),including vehicular safety, different sensors and real-time control. In 5G wireless communication networks, access nodes and / or UEs may havemultiple radio interfaces, such as below 6 gigahertz (GHz), centimeter wave (cmWave)and millimeter wave (mmWave), and also being integrable with legacy radio accesstechnologies, such as LTE. Integration with LTE may be implemented, for example, as a system, where macro coverage may be provided by LTE, and 5G radio interface access may come from small cells by aggregation to LTE. In other words, a 5G wireless communication network may support both inter-RAT operability (such asinteroperability between LTE and 5G) and inter-RI operability (inter-radio interfaceoperability, such as between below 6GHz, cmWave, and mmWave).5G wireless communication networks may also apply network slicing, inwhich multiple independent and dedicated virtual sub-networks (network instances)may be created within the same physical infrastructure to run services that havedifferent requirements on latency, reliability, throughput and mobility. In one embodiment, an access node 104 may comprise: a radio unit (RU) 103comprising a radio transceiver (TRX), i.e., a transmitter (Tx) and a receiver (Rx); one or more distributed units (DUs) 105 that may be used for the so-called Layer 1 (L1) processing and real-time Layer 2 (L2) processing; and a central unit (CU) 108 (also known as a centralized unit) that may be used for non-real-time L2 and Layer 3 (L3) processing. The CU 108 may be connected to the one or more DUs 105 for example viaan F1 interface. Such an embodiment of the access node 104 may enable thecentralization of CUs relative to the cell sites and DUs, whereas DUs may be more distributed and may even remain at cell sites. The CU and DU together may also be referred to as baseband or a baseband unit (BBU). The CU and DU may also be comprised in a radio access point (RAP). The CU 108 may be a logical node hosting radio resource control (RRC), service data adaptation protocol (SDAP) and / or packet data convergence protocol(PDCP), of the NR protocol stack for an access node 104. The CU 108 may comprise acontrol plane (CU-CP), which may be a logical node hosting the RRC and the control plane part of the PDCP protocol of the NR protocol stack for the access node 104. The CU 108 may further comprise a user plane (CU-UP), which may be a logical node hosting the user plane part of the PDCP protocol and the SDAP protocol of the CU for the access node 104. The DU 105 may be a logical node hosting radio link control (RLC), medium access control (MAC) and / or physical (PHY) layers of the NR protocol stack for the accessnode 104. The operations of the DU 105 may be at least partly controlled by the CU 108.It should also be understood that the distribution of functions between the DU 105 andthe CU 108 may vary depending on the implementation.Cloud computing systems may also be used to provide the CU 108 and / or DU 105. A CU provided by a cloud computing system may be referred to as a virtualized CU (vCU). In addition to the vCU, there may also be a virtualized DU (vDU) provided by a cloud computing system. Furthermore, there may also be a combination, where the DU may be implemented on so-called bare metal solutions, for example application-specific integrated circuit (ASIC) or customer-specific standard product (CSSP) system-on-a-chip (SoC). Edge cloud may be brought into the radio access network by utilizing network function virtualization (NFV) and software defined networking (SDN). Using edge cloud may mean access node operations to be carried out, at least partly, in acomputing system operationally coupled to a remote radio head (RRH) or a radio unit(RU) 103 of an access node 104. It is also possible that access node operations may beperformed on a distributed computing system or a cloud computing system located atthe access node 104. Application of cloud RAN architecture enables RAN real-timefunctions being carried out at the radio access network (e.g., in a DU 105), and non-real-time functions being carried out in a centralized manner (e.g., in a CU 108). 5G (or new radio, NR) wireless communication networks may support multiple hierarchies, where multi-access edge computing (MEC) servers may be placed between the core network 110 and the access node 104. It should be appreciated that MEC may be applied in LTE wireless communication networks as well. A 5G wireless communication network (“5G network”) may also comprise a non-terrestrial communication network, such as a satellite communication network, to enhance or complement the coverage of the 5G radio access network. For example, satellite communication may support the transfer of data between the 5G radio accessnetwork and the core network 110, enabling more extensive network coverage. Possibleuse cases may include: providing service continuity for machine-to-machine (M2M) orInternet of Things (IoT) devices or for passengers on board of vehicles, or ensuring service availability for critical communications, and future railway, maritime, or aeronautical communications. Satellite communication may utilize geostationary earth orbit (GEO) satellite systems, or low earth orbit (LEO) satellite systems, such as mega- constellations (i.e., systems in which hundreds of (nano)satellites are deployed).Alternatively, the satellites may be an airborne devices, such as an unmanned aerialvehicle (UAV), or a high-altitude platform system (HAPS). A given satellite 106 mayprovide communication services on Earth via one or more satellite beams. The one or more satellite beams create one or more cells over a given service area that may be bounded by the field of view of the satellite 106. It is obvious for a person skilled in the art that the access node 104 depictedin FIG. 1 is just an example of a part of a radio access network, and in practice the radioaccess network may comprise a plurality of access nodes 104, the UEs 100, 102 may haveaccess to a plurality of radio cells, and the radio access network may also comprise other apparatuses, such as physical layer relay access nodes or other entities. At least one of the access nodes may be a Home eNodeB or a Home gNodeB. A Home gNodeB or a Home eNodeB is a type of access node that may be used to provide indoor coverage inside a home, office, or other indoor environment. Additionally, in a geographical area of a radio access network, a plurality of different kinds of radio cells as well as a plurality of radio cells may be provided. Radio cells may be macro cells (or umbrella cells) which may be large cells having a diameterof up to tens of kilometers, or smaller cells such as micro-, femto- or picocells. The accessnode(s) 104 of FIG. 1 may provide any kind of these cells. A cellular radio network maybe implemented as a multilayer access networks including several kinds of radio cells. In multilayer access networks, one access node may provide one kind of a radio cell or radio cells, and thus a plurality of access nodes may be needed to provide such a multilayer access network. For fulfilling the need for improving performance of radio access networks,the concept of “plug-and-play” access nodes may be introduced. A radio access network,which may be able to use “plug-and-play” access nodes, may include, in addition to HomeeNodeBs or Home gNodeBs, a Home Node B gateway (HNB-GW) (not shown in FIG. 1).An HNB-GW, which may be installed within an operator’s radio access network, may aggregate traffic from a large number of Home eNodeBs or Home gNodeBs back to a corenetwork 110 of the operator.6G wireless communication 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 may 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. Artificial intelligence (AI) and machine learning (ML) may be used to enhancevarious aspects of wireless communication networks, such as network optimization,beam management, resource allocation, fault detection, maintenance, security, and / oruser experience. For example, in beam management, AI and ML algorithms may be usedboth at the UE side and at the network side to improve beamforming, beam tracking,and / or beam selection, which may be beneficial for maintaining optimal signal strengthand quality in high-frequency bands like millimeter-wave (mmWave). The integration ofAI and ML in wireless communication networks may result in increased efficiency,adaptability, and scalability. In the context of AI and ML, data collection is a process for gathering data bynetwork nodes, management entities, and / or UE(s), for the purpose of training anartificial intelligence or machine learning model, and / or for data analytics and inference.An artificial intelligence or machine learning model is a data-driven algorithmthat applies artificial intelligence or machine learning techniques to generate a set ofoutputs based on a set of inputs. The training is a process for training an artificial intelligence or machinelearning model by learning the relationship between inputs and outputs in a data-drivenmanner to obtain the trained model to be used for inference.The inference involves using a trained AI or ML model to produce a set ofoutputs (predictions) based on a set of inputs.AI or ML model validation is a subprocess of the training, which evaluates thequality of an AI or ML model using a dataset different from the one used for the modeltraining, helping to select model parameters that generalize beyond the training dataset. AI or ML model testing is another subprocess of the training, which evaluatesthe performance of a final AI or ML model using a dataset different from those used formodel training and validation. Unlike the model validation, testing does not assumesubsequent tuning of the model. AUE-side AI or ML model is one whose inference is performed entirely at theUE, while a network-side AI or ML model is one whose inference is performed entirelyat the network (e.g., at a gNB).A one-sided AI or ML model refers to either a UE-side or network-side AI orML model. In contrast, a two-sided AI or ML model involves paired models over whichjoint inference is performed, where the inference is conducted jointly across the UE and the network. This means the first part of the inference is performed by the UE, and the remaining part is performed by the gNB, or vice versa. Channel estimation is a process that involves determining the characteristicsof a wireless communication channel, such as the path loss, fading, complex valuecoefficients (i.e., phase and amplitude information), and / or interference. The channelestimate information may be used by the receiver to accurately decode the received datasignal (radio signal). The phase and amplitude information may also be referred to aschannel state information (CSI). Areference signal (RS), such as a demodulation reference signal (DMRS) or achannel state information reference signal (CSI-RS), may be embedded within thetransmitted data for the purpose of channel estimation. Reference signal resources aresignals associated with sequences (e.g., Pseudo-random or Zadoff-Chu or M-Sequences)that are known to both the transmitter and receiver, providing a reliable basis forchannel estimation. When the receiver detects these reference signals, it uses them tomeasure the channel’s phase and amplitude information (i.e., CSI), enabling the receiverto compensate for any distortions or variations in the data signal caused by the transmission environment. This process is beneficial for maintaining the integrity andquality of the communication link, for example in scenarios involving high mobility orcomplex propagation conditions. In MIMO systems, reference signals may be used forestimating the CSI for each logical antenna port (orthogonal or non-orthogonal) that isassociated with reference signal resources (or transmission path), facilitating spatialmultiplexing and beamforming for transmission of data and / or control information.The physical downlink shared channel (PDSCH) demodulation referencesignal (DMRS) may be used to assist UEs in demodulating and decoding data transmittedby the network (e.g., a gNB) on the PDSCH. The primary purpose of PDSCH DMRS is toprovide a known antenna port(s) associated with reference signal resource(s) that theUE can use to estimate the downlink CSI (i.e., effective downlink channel covering theimpact of DL transmission precoding, radio channel and receiver for the DMRSreception). PDSCH DMRS is also beneficial in MIMO communication by aiding in spatialmultiplexing, allowing the UE to distinguish orthogonal and non-orthogonal antennaports of DMRS transmitted via different physical transmit antenna elements at the gNB.The DMRS antenna ports may be positioned within the resource elements of one or morephysical resource blocks (PRBs) of the PDSCH resources (e.g., in time and / or frequencydomain), with its exact location varying depending on the numerology and transmissionconfiguration. PDSCH is physical channel used for transmitting user data in downlink(from the network to the UE 100). The sequence for DMRS may be generated by using a specific sequence type(e.g., Pseudo-Random or M-Sequence or Zadoff-Chu) based on a pre-defined patternknown to both the transmitter and receiver, thus facilitating accurate estimation of thechannel or CSI. The DMRS sequence may be modulated using schemes such asquadrature phase shift keying (QPSK) or quadrature amplitude modulation (QAM). Aftermodulation, these DMRS symbols may be mapped onto a specific resource elementpattern within one or more PRBs associated with the PDSCH.For example, the DMRS-DownlinkConfig information element may beincluded in an RRC configuration to configure downlink demodulation reference signalsfor PDSCH. In general, DMRS may be associated with specific antenna ports withcorresponding resource elements. DMRS antenna ports are logical entities that help thenetwork distinguish channels (i.e., effective channel) or CSI between PDSCH layerstransmitted throughout multiple transmit antennas. A resource element (RE) is a unit of resource allocation in the time-frequencygrid of wireless communication systems associated with applied subcarrier spacing (e.g.,15 or 30 or 60 or 120 or 240 kilohertz). For example, a resource element may representa single subcarrier in the frequency domain and / or a single OFDM symbol in the timedomain. A given resource element may carry a piece of data or control information or areference signal. The DMRS can be transmitted in various schemes, including time-domain,frequency-domain, and beamformed (covering also precoding) DMRS, each suited todifferent transmission scenarios. Upon receiving the DMRS symbols, the UE may use theDMRS symbols to estimate the channel conditions or CSI for each antenna port (AP) orbeam, which is beneficial for coherent demodulation (covering potentially alsointerference cancellation or mitigation) and decoding of the PDSCH transport block(s).At the UE side, DMRS antenna ports enable the receiver to accurately estimate thechannel (i.e., effective channel) or CSI associated with the transmitted PDSCH layer(s) byknowing the exact reference signal configuration, which may include, for example, theantenna port mapping in frequency and time. After the UE has computed the AP-specificDMRS channel estimates based on the DMRS configuration, the UE may use thedetermined channel estimates for the demodulation of the PDSCH.For example, TS 38.211 (version 18.0.0) defines the following configurationsfor DMRS mapping to physical resources. The UE may assume the PDSCH DMRS beingmapped to physical resources according to configuration type 1 or configuration type 2as given by the higher-layer parameter dmrs-Type. The UE may assume the sequencer(m) is scaled by a factor ^ DMRSPDSCH to conform with the transmission power specified in TS38.214 and mapped to resource elements (^, ^)^,^ according to, if the higher-layerparameter dmrs-TypeEnh is configured: ^^^^,^^^,^= ^PDDMSRCSH ^f(^^)^t(^^)^(4^ + ^^)8^ + 2^^ + Δ configuration type 1^ = ^ 12^ + ^^ + Δ configuration type 2, ^′ = 0,112^ + ^^ + Δ + 4 configuration type 2, ^′ = 2,3^^ = 0,1,2,3^ = ^ ̅ + ^′^ = 0,1, …^ = 0,1, … , ^ − 1Otherwise: configuration type 1 configuration type 2^^ = 0,1^ = ^ ̅ + ^′^ = 0,1, …^ = 0,1, … , ^ − 1where ^f(^^), ^ (^^t ), and Δ are given by Tables 7.4.1.1.2-1 and 7.4.1.1.2-2 of TS 38.211,and the following conditions are fulfilled: the resource elements are within the commonresource blocks allocated for PDSCH transmission. The reference point for ^ and the position ^^ of the first DMRS symbol dependon the mapping type. For PDSCH mapping type A, ^ is defined relative to the start of theslot, with ^^ = 3 if the higher-layer parameter dmrs-TypeA-Position is ‘pos3’ and ^^ = 2otherwise. For PDSCH mapping type B, ^ is defined relative to the start of the scheduledPDSCH resources, with ^^ = 0.The position(s) of the DMRS symbols are given by ^ ̅ and the duration ^^ . ForPDSCH mapping type A, ^^ is the duration between the first orthogonal frequency-division multiplexing (OFDM) symbol of the slot and the last OFDM symbol of thescheduled PDSCH resources in the slot. For PDSCH mapping type B, ^^ is the duration ofthe scheduled PDSCH resources. The time-domain index ^′ and the supported antenna ports ^ are given byTable 7.4.1.1.2-5 of TS 38.211. Single-symbol DMRS may be used if the higher-layerparameter maxLength in the DMRS-DownlinkConfig information element is notconfigured. Single-symbol or double-symbol DMRS may be determined by the associateddownlink control information (DCI) if the higher-layer parameter maxLength is equal to‘len2’. Basic or enhanced DMRS multiplexing may be controlled by the higher-layerparameter dmrs-TypeEnh. Table 1 provides an example of type-1 and type-2 DMRS resource element pattern resource overhead in frequency domain over PDSCH allocation per symbol,when all DL DMRS antenna ports are indicated or configured for the UE. In Table 1, “FL”is an abbreviation for “front-loaded”, which refers to the first DMRS symbol before PDSCH reception. DMRS frequency domain overhead (%) 1FL Type-1 1 FL Type-2100 % 100 %Table 1. Type-1 and Type-2 DMRS resource overhead in frequency domainper symbol when all DMRS antenna ports are indicated for UE(s). As shown in Table 1, when all DL DMRS antenna ports are indicated for UE(s),DMRS resource overhead in frequency domain may go up to 100%, leading to very highDMRS resource overhead in frequency domain. It should be noted that, when themaximum number of DMRS antenna ports is configured for the UE, for example 8 APs(type-1), 12 APs (type-2), or 24 APs (e-type2), the current NR specification requires thatDL DMRS is allocated fully over the PDSCH allocation (i.e., covering all physical resourceblocks and related resource elements assigned for PDSCH bandwidth). In other words,the current NR specification does not support any technique to reduce the frequencydensity of DMRS. Some example embodiments may utilize a channel predictor (e.g., an AI-based or ML-based model) for a downlink reference signal (e.g., PDSCH DMRS or anyother downlink reference signal). The channel predictor may provide predictions of thechannel conditions (i.e., effective channel) or CSI in frequency and / or time domain foreach antenna port associated with the downlink reference signal. Thus, the UE does not need to receive the actual downlink reference signal on the resource elements (infrequency and / or time domain) associated with the predicted channel estimates (andthe network does not need to transmit the actual downlink reference signal on those resource elements). Instead, the UE can rely on the predictions made by the channelpredictor to estimate the channel conditions (in frequency and / or time domain) forthose resource elements. This approach can reduce the overhead associated with transmittingreference signals and improve the overall efficiency of the communication system. Inother words, the example embodiments described herein may enable resource overhead reduction in frequency and / or time domain for example for PDSCH DMRS (or any other downlink reference signal). The example embodiments enable flexible usage of the reference signal resource element patterns for the resource overhead reduction. For example, the channel predictor may be trained with a specific physical resource block (PRB) level granularity associated with a certain reference signal resource element pattern (e.g., DMRS type-1, e-type1, 6G-type, etc.) in frequency domain,and / or symbol position pattern (e.g., every n-th symbol or symbol indices 4, 7, 10 out of14 symbol in a slot) in time domain, a specific number of antenna ports, a specific sequence type (e.g., pseudo-random) with one or more specific initialization values (e.g.,initialization seed values), and a specific downlink precoding technique.Some example embodiments are described below using principles and terminology of 5G radio access technology without limiting the example embodiments to 5G radio access technology, however. FIG. 2A illustrates the training phase 201 of a machine learning model 210(channel predictor) used for channel prediction associated with channel estimation. Thetraining may be performed in a UE 100, 102, for example.As an example, the machine learning model 210 may be or comprise a convolutional neural network (CNN). CNNs are designed to process grid-like data,similar to the channel (effective channel) or CSI matrix, wherein the CNN can extractfeatures of the channel. CNNs can also capture temporal and / or frequency domain correlations in channel variations. Therefore, a CNN model can be applied for predictionof the channel (effective channel) or CSI, such that the CNN model predicts non-transmitted resource elements associated with DMRS antenna ports by leveragingmeasurements associated with transmitted resource elements associated with DMRSantenna ports. As another example, the machine learning model 210 may be or comprise asupervised learning model. However, it should be noted that the machine learning model 210 is not limited to these examples, and any other suitable type of machine learning model can be used. The machine learning model 210 is trained based on a set of input data 211and a set of expected output data 212. In the training phase 201, the machine learningmodel 210 learns the relationship or logic between the set of input data 211 and theexpected output data 212. During the training phase 201, all reference signal resource elements (e.g.,DMRS REs) with a specific resource element pattern (e.g., DMRS e-type1 or e-type2)associated with reference signal antenna ports (e.g., DMRS APs) across the configuredPDSCH bandwidth may be transmitted by the network (e.g., network node 104) with aconfigured set of symbols (e.g., all symbols in a slot or a smaller set) in time domain. TheUE 100 may compute or determine corresponding antenna-port-specific channelestimates across the configured PDSCH bandwidth and configured set of symbols in time domain. The set of input data 211 of the machine learning model 210 may comprise,per antenna port associated with the downlink reference signal, one or more channelestimates (in at least one of the frequency domain or the time domain) associated with afirst set of received resource elements of the downlink reference signal (e.g., PDSCHDMRS). The set of input data 211 further comprises a downlink reference signalconfiguration associated with the downlink reference signal (i.e., the necessaryinformation about the configuration and characteristics of the downlink reference signal). The set of expected output data 212 of the machine learning model 210 may comprise one or more channel estimates (in at least one of the frequency domain or thetime domain) per antenna port associated with a second set of received resourceelements of the downlink reference signal (e.g., PDSCH DMRS). The expected output data 212 teaches the machine learning 210 about what the output of the machine learning model 210 (i.e., the channel predictions to be made by the ML model) should be. The downlink reference signal configuration may comprise, for example, atleast one of the following information elements: a type of the downlink reference signal,a physical resource block level granularity associated with a resource element patterntype of the downlink reference signal in frequency domain, a reference signal symbol(e.g., DMRS symbol) position pattern in time domain, a number of antenna portsassociated with the downlink reference signal, a sequence type of the downlinkreference signal with one or more initialization seed values, and / or a downlinkprecoding technique of the downlink reference signal. These information elements maybe identified by one or more data set identifiers (dataSetIDs).The type of the downlink reference signal refers to the specific configuration or variant of the reference signal used in the downlink transmission. Different types of reference signals are designed to meet various requirements and scenarios in wireless communication systems. Each type of downlink reference signal is associated with specific characteristics, such as the pattern of resource elements it occupies, the sequence used for the signal, and the way it is mapped to physical resources. Someexamples of the type of the downlink reference signal may include (but are not limitedto): DMRS Type-1, DMRS Type-2, DMRS E-Type 1 (Enhanced Type-1), or DMRS 6G-Type.DMRS Type-1 is a standard type of DMRS used in 5G NR (e.g., used for channel estimationin scenarios with lower mobility and simpler propagation conditions). DMRS Type-2may be used for more complex scenarios, such as those involving multiple antenna ports in MIMO systems. E-Type1 is an enhanced version of the standard DMRS Type-1, which may include additional features or modifications to improve performance in more challenging conditions, such as higher mobility or more complex propagationenvironments. The 6G-Type refers to a type of reference signal that might be used infuture 6G networks. The symbol position pattern refers to the specific configuration in timedomain where a set of reference signal symbol (e.g., DMRS symbol) positions in timedomain within a slot for reference signal transmission (e.g., DMRS transmission) aredefined. In one example, the symbol positions may be defined as an offset with respectto a specific control symbol, data symbol or reference signal symbol within a slot, suchthat the offset is always larger than zero. In another example, the symbol positionpattern may be an absolute symbol index from the start of the slot. In another example,the symbol position pattern may define the time density of reference signal symbols (e.g.,DMRS symbols) within a slot, starting from a specific symbol position till the end of theslot. The physical resource block level granularity refers to the detailed allocation of resource elements within the PRBs, which can be specified in the frequency domain. The physical resource block level granularity indicates how the resources are distributed and used for the downlink reference signal. The number of antenna ports associated with the downlink reference signalspecifies the number of antenna ports that represent the transmission paths of thedownlink reference signal, through which the downlink reference signal is transmittedfrom the network (e.g., gNB) to the UE. In other words, an antenna port (e.g., DMRS port)is a logical representation of a transmission path used to transmit the downlinkreference signal (e.g., DMRS) from the network to the UE (i.e., the antenna port does notrefer to a physical antenna). Each antenna port corresponds to a different transmission path, and knowing the number of antenna ports helps in understanding the spatialconfiguration of the downlink reference signal. The downlink reference signal may bemapped to one or more antenna ports. The UE may then use the downlink referencesignal received on these antenna port(s) to estimate the channel conditions or CSI oreffective channel for each transmission path.The “sequence type of the downlink reference signal with one or moreinitialization seed values” includes the type of sequence used for the downlink referencesignal (e.g., pseudo-random sequence) and the initialization seed value(s) that are usedto generate the sequence. The sequence type and seed values may be needed forsynchronizing the transmitter and receiver. The downlink precoding technique refers to the technique used to pre-process the downlink reference signal before transmission to improve signal quality andreduce interference. For example, precoding may involve applying a set of weights to thetransmitted signal across multiple antennas to optimize the signal reception at thereceiver. Precoding may be codebook-based or non-codebook-based.FIG. 2B illustrates the inference phase 202 of the trained machine learningmodel 220, where the trained machine learning model 220 is used to make channelpredictions associated with channel estimation (after the training phase 201 of FIG. 2Ais completed). During the inference phase 202, the network node 104 may transmit areduced amount of PRBs associated with reference signal resource elements (e.g., DMRSREs) of a specific resource element pattern (e.g., DMRS e-type1 or e-type2) associatedwith reference signal antenna ports (e.g., DMRS APs) over a reduced set of configuredsymbols. The UE 100 may perform measurements and determine corresponding channelestimates based on the reduced density PRBs and reduced set of symbols transmitted bythe network node 104. The UE 100 may feed this information into the trained machinelearning model 220, which then predicts antenna-port-specific channel estimates on theresource elements and PRBs which have not been transmitted by the network node 104(i.e., on the “gaps” on which the UE 100 does not receive the reference signal). The non-transmitted resource “gaps” may be in frequency and / or time domain.In the inference phase 202, the set of input data 221 of the trained machinelearning model 220 may comprise, per antenna port associated with the downlinkreference signal, one or more channel estimates (in at least one of the frequency domain or the time domain) associated with a set of received resource elements of the downlink reference signal. The set of input data 211 further comprises a downlink reference signalconfiguration associated with the downlink reference signal (i.e., the necessaryinformation about the configuration and characteristics of the downlink reference signal). Based on the set of input data 221, the trained machine learning model 220generates output data 222 comprising one or more predicted channel estimates perantenna port associated with the downlink reference signal. The one or more predictedchannel estimates are associated with one or more non-received resource elements ofthe downlink reference signal (i.e., the UE does not need to actually receive the downlinkreference signal on the resource elements, for which the channel estimates are predicted). The “one or more predicted channel estimates per antenna port” means thatthe trained ML model 220 provides one or more predictions of the channel conditionsfor each logical transmission path (antenna port) associated with the downlink referencesignal. FIG.3 illustrates a signal flow diagram according to an example embodiment. Referring to FIG. 3, at 301, a UE 100 transmits a capability indication to a network node (access node) 104 of a radio access network, wherein the capabilityindication indicates that the UE 100 supports channel prediction associated with channelestimation. For example, the capability indication may be comprised in a radio resourcecontrol (RRC) message generated by the UE 100. The network node 104 receives thecapability indication. The network node 104 may refer to a base station (e.g., a gNB) controlling a serving cell of the UE 100 (i.e., a cell that the UE 100 is connected to). The capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation. The ”channel prediction associated with channel estimation” means that theUE 100 is capable of using the channel predictor to predict channel estimates (withoutreceiving an actual reference signal on the resource elements for which the channelestimates are predicted by the channel predictor). The channel predictor associated withthe channel estimation means that the channel predictor is used in conjunction with the channel estimation process. For example, the channel predictor may refer to the trainedmachine learning model 220 described above. Alternatively, the channel predictor maybe a non-ML (non-AI) model, or based on a combination of ML and non-ML inference.The one or more data set identifiers specify the capabilities of the channelpredictor. A given data set identifier may identify a set or subset of reference signalconfiguration data used for training the channel predictor.For example, one of the one or more data set identifiers may indicate at leastone physical resource block density allocation of a reference signal resource elementpattern type used for training the channel predictor (e.g., at resource element and PRBlevel). In other words, the channel predictor may have been trained with a specific PRB-level granularity associated with a certain resource element pattern (e.g., DMRS type-1,e-type1, 6G-type, etc.) in frequency domain, and a specific data set identifier may be usedto define which PRB density allocation has been used for training the channel predictor(i.e., as an input of the channel predictor).Alternatively, the one or more data set identifiers may indicate at least one physical resource block density allocation of a reference signal resource element patterntype that the channel predictor is capable of predicting (which may be the same ordifferent than the PRB density allocation used for training the channel predictor). Inother words, in this approach, the capability indication may indicate which PRB densityallocation the trained channel predictor model can predict as an output of the channelpredictor. Alternatively, the one or more data set identifiers may indicate at least onesymbol position pattern that the channel predictor is capable of predicting in timedomain (which may be the same or different than the symbol position pattern used for training the channel predictor). In other words, in this approach, the capability indication may indicate which symbol position pattern the trained channel predictor model can predict in time as an output of the channel predictor. Some examples of the PRB density allocation may include, but are not limitedto: an even PRB density allocation (see FIG. 4), an odd PRB density allocation (see FIG.4), an irregular PRB density allocation (see FIG. 4), or every n-th PRB density allocation,where n>1. For example, when the channel predictor has been trained with the “even PRBdensity allocation” option, it means that the channel predictor can predict antenna portsof resource elements associated with odd-numbered PRBs. In other words, in this case,the model training has been done with even PRBs and corresponding antenna ports andresources therein, and thus the channel predictor can predict the channel estimatesassociated with odd-numbered PRBs (which are not transmitted by the network node104). The channel predictor may be trained with a specific reference signal resourceelement pattern type or combination of resource element pattern type (e.g., DMRS type1, or 6G-type1, or 6G-type1 + type2, or any other 6G-type(s)), such that every n-th oddPRB comprises a reference signal resource element pattern (or combination of patterns)associated with reference signal antenna ports (e.g., DMRS antenna ports) of a downlinkdata channel’s frequency domain allocation (e.g., PDSCH frequency domain allocation). A non-limiting example of a data set identifier for the “even PRB density allocation”option may be DMRS-typeX-FreqDensityEvenDataSetID#n, where X refers to the specificreference signal resource element pattern type (e.g., DMRS type 1, or 6G-type1, or 6G- type1 + type2, or any other 6G-type(s)). It is also possible to have different data setidentifiers for different values of n (the larger the value of n, the sparser the allocation ofREs for the reference signal). As another example, when the channel predictor has been trained with the“odd PRB density allocation” option, it means that the channel predictor can predictantenna ports of resource elements associated with even-numbered PRBs. In otherwords, in this case, the model training has been done with odd PRBs and correspondingantenna ports and resources therein, and thus the channel predictor can predict thechannel estimates associated with even PRBs (which are not transmitted by the networknode 104). The channel predictor may be trained with a specific reference signalresource element pattern type or combination of resource element pattern types (e.g.,DMRS type 1, or 6G-type1, or 6G-type1 + type2, or any other 6G-type(s)), such that everyj-th even PRB comprises a reference signal resource element pattern (or combination ofpatterns) associated with reference signal antenna ports (e.g., DMRS antenna ports) of a downlink data channel’s frequency domain allocation (e.g., PDSCH frequency domainallocation). A non-limiting example of a data set identifier for the “odd PRB densityallocation” option may be DMRS-typeX-FreqDensityOddDataSetID#j, where X refers tothe specific reference signal resource element pattern type (e.g., DMRS type 1, or 6G-type1, or 6G-type1 + type2, or any other 6G-type(s)). It is possible to have different dataset identifiers for different values of j (the larger the value of j, the sparser the allocationof REs for the reference signal). In the “irregular PRB density allocation” option, the channel predictor may be trained with a specific reference signal resource element pattern type or combination of resource element pattern types (e.g., DMRS type 1, or 6G-type1, or 6G-type1 + type2, orany other 6G-type(s)), such that a combination of every n-th odd and every j-th even PRBcomprises a reference signal resource element pattern (or combination of patterns)associated with reference signal antenna ports (e.g., DMRS antenna ports) of a downlinkdata channel’s frequency domain allocation (e.g., PDSCH frequency domain allocation).A non-limiting example of a data set identifier for the “irregular PRB density allocation”option may be DMRS-typeX-FreqDensityEvenDataSetID#n + DMRS-typeX-FreqDensityOddDataSetID#j, where X refers to the specific reference signal resource element pattern type (e.g., DMRS type 1, or 6G-type1, or 6G-type1 + type2, or any other 6G-type(s)). As another example, when the channel predictor has been trained with the “symbol position pattern with time density” option, it means that the channel predictor can predict antenna ports of resource elements associated with configured PRB densitywith configured symbol position pattern with a specific time density. In other words, inthis case, the model training has been done with reference signal symbols (e.g., DMRSsymbols) with a specific time density of reference signal symbols (e.g., DMRS symbols)and corresponding antenna ports and resources therein, and thus the channel predictor can predict the channel estimates associated with symbol position pattern with density (which are not transmitted by the network node 104). The channel predictor may be trained with a specific symbol position pattern such that configured reference signalsymbols (e.g., DMRS symbols) comprise a reference signal resource element pattern (orcombination of patterns) associated with reference signal antenna ports (e.g., DMRS antenna ports) of a downlink data channel’s frequency domain allocation (e.g., PDSCHfrequency domain allocation) as well as symbol pattern in time domain. A non-limitingexample of a data set identifier for the “symbol position pattern with offset” option may be DMRS-typeX-SymbolPositionPattern-offset#k-DataSetID#j, where X refers to the specific reference signal resource element pattern type (e.g., DMRS type 1, or 6G-type1, or 6G-type1 + type2, or any other 6G-type(s)). It is possible to have different data setidentifiers for different values of k (the larger the value of k, the sparser the allocation ofDMRS symbols in time for the reference signal).Alternatively, or additionally, the one or more data set identifiers mayindicate at least one of: one or more carrier frequencies supported by the channelpredictor, one or more numerology options supported by the channel predictor percarrier frequency, one or more delay spread ranges (in time) supported by the channelpredictor, one or more Doppler frequency shift or spread values supported by thechannel predictor, one or more UE speed values supported by the channel predictor, oneor more reference signal sequence initialization seed values (or a range of values)supported by the channel predictor, a number of reference signal antenna portssupported by the channel predictor, one or more reference signal resource element typessupported by the channel predictor, one or more reference signal sequence typessupported by the channel predictor, a physical downlink shared channel allocationlength (e.g., in PRBs and / or in symbols) or range (in at least one of the frequency domainor the time domain) supported by the channel predictor, one or more precoding typessupported by the channel predictor, or a precoding granularity (in at least one of thefrequency domain or the time domain) supported by the channel predictor.For example, each of the above information elements may be indicated by aseparate data set identifier, or a single data set identifier may indicate a certaincombination of these information elements. In other words, the capability indicationmay indicate one or more information elements per each data set identifier, or multiple(a group of) data set identifiers may share part of the information elements with eachother. Acarrier frequency refers to the frequency at which the reference signal istransmitted. By indicating the one or more supported carrier frequencies in thecapability indication, the network node 104 can be made aware of which carrierfrequencies can be configured for the trained channel predictor of the UE 100. As a non-limiting example, the supported carrier frequencies may be 700 MHz to 3.5 GHz and 7 to15 GHz. The one or more numerology options refer to the different numerologyconfigurations (e.g., subcarrier spacing) used for the reference signal. By indicating theone or more supported numerology options in the capability indication, the networknode 104 can be made aware of which subcarrier spacing options can be configured forthe trained channel predictor of the UE 100. As a non-limiting example, the one or moresupported numerology option may be indicated with L-bit quantized format (e.g., withtwo bits: ‘00’ for 15 KHz, ‘01’ for 30 KHz, ‘10’ for 60KHz, ‘11’ for 120KHz).The one or more delay spread ranges refer to the time dispersion of thereference signal. By indicating the one or more supported delay spread ranges in thecapability indication, the network node 104 can be made aware of whether the trainedchannel predictor of the UE 100 can be utilized in the presence of a certain delay spread.The network node 104 may have this awareness based on uplink reference signalmeasurements. As a non-limiting example, the one or more supported delay spreadranges may be indicated with N-bit quantized format (e.g., with two bits indicating fourdifferent delay spread values: ‘00’ for 0-10 ns, ‘01’ for 11-30 ns, ‘10’ for 31-100 ns, ‘11’for 101- 300ns).The one or more Doppler frequency shift or spread values refer to the timevariation of the reference signal. By indicating the one or more supported Dopplerfrequency shift or spread ranges in the capability indication, the network node 104 canbe made aware of whether the trained channel predictor of the UE 100 can be utilized inthe presence of a Dopper frequency shift or spread. The network node 104 may have thisawareness based on uplink reference signal measurements. The one or more UE speed values refer to the time variation of the referencesignal. By indicating the one or more supported UE speed values or ranges in the capability indication, the network node 104 can be made aware of whether the trained channel predictor of the UE 100 can be utilized in the presence of UE mobility with acertain UE speed or Doppler frequency shift or Doppler spread. The network node 104may have this awareness based on uplink reference signal measurements. The one or more reference signal sequence initialization seed values refer to the initial value(s) used to generate the sequence (e.g., pseudorandom sequence) for thereference signal. By indicating the one or more supported initialization seed values inthe capability indication, the network node 104 can be made aware of which initialization values (e.g., DMRS initialization values) can be configured for the trainedchannel predictor of the UE 100. As a non-limiting example, the one or more supportedinitialization seed values may be indicated with K-bit quantized format (e.g., ‘00’ for seedvalues 0-8000, ‘01’ for seed values 8000-16000, ‘10’ for seed values 16000-24000, ‘11’ for seed values 24000-32000). The number of antenna ports defines the number of antenna ports (i.e., thelogical entities) used to transmit the reference signal. By indicating the supportednumber of antenna ports in the capability indication, the network node 104 can be madeaware of up to which total number of reference signal antenna ports (e.g., DMRS antennaports) the trained channel predictor of the UE 100 can be configured with. For example,the supported number of antenna ports may be 12, 24, 48 or any other number. In analternative approach, the supported number of antenna ports may indicate how manyadditional reference signal antenna ports (e.g., additional DMRS antenna ports) thechannel predictor can predict (i.e., this would be a smaller number compared to the totalnumber of reference signal antenna ports). This may also be subject to the referencesignal type (e.g., DMRS type), for example, due to different RE patterns associated withdifferent types, this capability may be DMRS-type-specific. The one or more reference signal resource element types refer to the specificways that the reference signal is mapped within the time-frequency grid of the radioframe. For example, there may be at least two different types of DMRS resourceelements: Type 1 and Type 2. Type 1 (up to 2 antenna ports) uses every second resourceelement within the symbols allocated to DMRS, effectively utilizing 50% of the availableresource elements. Type 2 (up to 2 antenna ports) uses every third resource elementwithin the symbols allocated to DMRS, utilizing about 33% of the available resource elements.The one or more reference signal sequence types refer to the types ofsequences (e.g., pseudo-random, Zadoff-Chu, m-sequence, or any other sequence or acombination of sequences) used for the reference signal. By indicating the one or moresupported sequence types in the capability indication, the network node 104 can bemade aware of which sequence type(s), the trained channel predictor of the UE 100 canbe configured with. The PDSCH allocation length or range refers to the length or range of thePDSCH allocation in either the frequency domain or the time domain, or in both thefrequency domain and the time domain. By indicating the supported PDSCH allocationlength or supported allocation range in frequency and / or time domain in the capabilityindication, the network node 104 can be made aware of up to which number of PDSCHPRBs and / or symbols the UE (or the trained channel predictor of the UE 100) can be configured with. Herein the number of PDSCH PRBs refers to the output of the channelpredictor in frequency domain, i.e., how wide PDSCH allocations in frequency domainthe channel predictor can be configured with (e.g., 52 PRBs, or 200 PRBs, or 1000 PRBs,etc.). PDSCH symbols refers to the output of the channel predictor in time domain, i.e.,what is the length of PDSCH allocation in time that the channel predictor can be configured with. The one or more precoding types refer to the types of precoding techniques(e.g., codebook-based, non-codebook-based) used to pre-process the reference signalbefore transmission. By indicating the one or more precoding types in the capabilityindication, the network node 104 can be made aware of which DL precodingtechnique(s) the trained channel predictor of the UE 100 can be configured with. Someexamples of precoding types may include (but are not limited to): codebook-based (e.g.,type-1, type2, enhanced type 2, further enhanced type 1), or non-codebook based.The precoding granularity refers to the level of detail at which the precodingis applied, such as per resource block, per subcarrier, or per symbol, to optimize signaltransmission and reception. By indicating the supported precoding granularity in thecapability indication, the network node 104 can be aware of which precoder resourceblock group (PRG) sizes for a downlink data channel (e.g., PDSCH) the trained channelpredictor of the UE 100 can be configured with. At 302, the network node 104 determines or generates, based on thecapability indication, a downlink reference signal configuration to be used for thechannel predictor of the UE 100, such that the downlink reference signal configurationis supported by the channel predictor of the UE 100. For example, the downlinkreference signal configuration may be determined such that it corresponds to the reference signal configuration used for training the channel predictor. The downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one of a frequency domain or a time domain. The transmission pattern and / or symbol position pattern comprises orindicates at least a first set of resource elements on which the network node 104 intendsto transmit the downlink reference signal. The transmission pattern and / or symbolposition pattern may further comprise or indicate a second set of resource elements onwhich the network node 104 does not intend to transmit the downlink reference signal.In other words, the second set of resource elements refers to resource elements forwhich the UE 100 is expected or configured to predict the channel (or effective channel)by using the channel predictor (without the network node 104 actually transmitting thedownlink reference signal and without the UE 100 actually receiving the downlinkreference signal on those resource elements and / or symbols).For example, the second set of resource elements may be explicitly indicated in the transmission pattern of the downlink reference signal configuration transmitted from the network node 104. Alternatively, the second set of resource elements may not be explicitlyindicated in the transmission pattern (i.e., in this case, the transmission pattern maycomprise or indicate only the first set of resource elements). In this case, the UE 100 maydetermine the second set of resource elements (for which the channel prediction is to beperformed) based on the first set of resource elements and / or the downlink referencesignal configuration used for training the channel predictor (e.g., based on the at leastone physical resource block density allocation of the reference signal resource element pattern type used for training the channel predictor). For example, if the channel predictor has been trained to predict the channel estimates associated with odd- numbered PRBs, then the UE 100 may determine or assume that the second set of resource elements corresponds to resource elements of the odd-numbered PRBs. Both the first set of resource elements and the second set of resourceelements may be comprised in physical resource blocks allocated for a downlink data channel (e.g., PDSCH) associated with the downlink reference signal (e.g., PDSCH DMRS). By not transmitting the downlink reference signal on the second set of resourceelements, the resource overhead of the downlink reference signal is reduced, such thatthe downlink reference signal is not allocated fully over the entire allocation of theassociated downlink data channel. I.e., the downlink reference signal is not spread acrossall the PRBs and related REs and / or configured symbols assigned for the downlink datachannel. The first set of resource elements may comprise at least one of: one or morefrequency resources (e.g., one or more subcarriers), and / or one or more time resources(e.g., one or more OFDM symbols). The second set of resource elements may comprise atleast one of: one or more frequency resources (e.g., one or more subcarriers), and / or oneor more time resources (e.g., one or more OFDM symbols). The first set of resourceelements and the second set of resource elements may be different such that they do notoverlap in time and / or frequency.The first set of resource elements may be associated with one or more first antenna ports that are equivalent to or different than one or more second antenna portsassociated with the second set of resource elements. In other words, the second set ofresource elements may be associated with the same antenna port(s) as the first set of resource elements, or the second set of resource elements may be associated withdifferent antenna port(s) than the first set of resource elements.. The one or more firstantenna ports and the one or more second antenna ports may be associated with datalayers of the downlink data channel associated with the downlink reference signal.The downlink reference signal configuration may further comprise at least one of: an indication for operating the channel predictor in at least one of the frequency domain or the time domain, or at least one data set identifier from the one or more data set identifiers to be applied for the channel predictor. Alternatively, the network node 104 may generate a separate configurationcomprising the at least one of: the indication for operating the channel predictor in atleast one of the frequency domain or the time domain, or the at least one data set identifier from the one or more data set identifiers to be applied for the channel predictor, and the network node 104 may transmit this separate configuration to the UE100 (e.g., via RRC or MAC signaling) in addition to the downlink reference signalconfiguration. At 303, the network node 104 transmits the downlink reference signalconfiguration to the UE 100. For example, the downlink reference signal configurationmay be transmitted via RRC signaling (e.g., in an RRC reconfiguration message) or MACsignaling. The UE 100 receives the downlink reference signal configuration.The network node 104 may also indicate the allocation of the downlink datachannel (e.g., PDSCH allocation) to the UE 100. For example, the allocation of thedownlink data channel may be indicated via one or more time domain resourceallocation (TDRA) tables, which define the scheduled resources of the downlink datachannel over time slots.A non-limiting example of the downlink reference signal configuration (e.g.,DL DMRS configuration) is provided in the following with four different data setidentifiers (it should be noted that a higher or lower number of data set identifiers thanfour may be used in an actual configuration). In this example, the “DMRS-DownlinkConfig” information element may be used to configure downlink demodulationreference signals for PDSCH, where the channel predictor can be configured to operatein the frequency domain and / or time domain with the higher layer parameter dmrs-predictor-domain as follows:-- ASN1START-- TAG-DMRS-DOWNLINKCONFIG-STARTDMRS-DownlinkConfig ::= SEQUENCE { --Void Text --- dmrs-predictor-domain = ENUMERATED {Frequency, Time, FrequencyAndTime} ---if dmrs-predictor-domain not configured following dataSetIDs are not valid dmrs-DataSetIds = SEQUENCE { dmrs-PredictorsSet1 = dataSetID1 – refers to dataSetIDindicated in capability signaling dmrs-PredictorsSet2 = dataSetID2 – refers to dataSetIDindicated in capability signaling dmrs-PredictorsSet3 = dataSetID3 – refers to dataSetIDindicated in capability signaling dmrs-PredictorsSet4 = dataSetID4 – refers to dataSetIDindicated in capability signaling } --Void text --- }-- TAG-DMRS-DOWNLINKCONFIG-STOP-- ASN1STOPAt 304, the UE 100 prepares the channel predictor based on the downlinkreference signal configuration. For example, in case the channel predictor comprises aCNN model, the preparation may include (but is not limited to) at least one of thefollowing: loading the trained model 220 from memory into a processing unit, initializingthe model with pre-trained weights that correspond to the trained model, preparingestimated channel or CSI estimate samples (estimated based on an actually receivedreference signal) into a suitable format (i.e., corresponding to the format used during thetraining), or normalizing and / or scaling the input of the trained model similarly asduring the model training. This way consistency can be ensured. At 305, the network node 104 transmits the downlink reference signal to theUE 100 on the first set of resource elements. For example, the downlink reference signalmay comprise (but is not limited to) one of: a physical downlink shared channel demodulation reference signal (PDSCH DMRS), a non-zero power (NZP) channel stateinformation reference signal (CSI-RS) for CSI acquisition or NZP-CSI-RS for time andfrequency tracking, or a phase-tracking reference signal (PTRS). The downlink reference signal is not transmitted on the second set of resource elements in at least one of the frequency domain or the time domain. In otherwords, the network node 104 may refrain from transmitting the downlink referencesignal on the second set of resource elements, so that the downlink reference signal istransmitted with reduced resource overhead. Consequently, the UE 100 does not receive the downlink reference signal onthe second set of resource elements. In other words, the UE 100 may be assumed to notreceive the second set of resource elements, for which the UE 100 shall perform antenna-port-specific channel prediction (i.e., the UE 100 may not receive any signal on thesecond set of resource elements). Alternatively, the UE 100 may assume that the downlink data channel’sresource allocation (e.g., PDSCH resource allocation) in frequency domain is mapped tothe second set of resource elements for which the channel prediction is performed (i.e.,the UE 100 may receive data from the network node 104 on the second set of resourceelements for which the channel prediction is performed). This option may help toincrease the capacity of the data channel such as PDSCH (i.e., increase the throughput ofthe data channel of a single user or data channels associated with multiple users) byenabling transmission of data associated with one or more users (or UEs) on thereference signal resource elements. At 306, the network node 104 transmits a downlink signal to the UE 100. TheUE 100 receives the downlink signal. The downlink signal may be transmitted on at leastone resource element of the second set of resource elements in at least one of thefrequency domain or the time domain. Thus, the second set of resource elements may beutilized to provide an enhanced multiplexing possibility for any downlink signal (insteadof using the second set of resource elements to transmit the downlink reference signal). In this case, the UE 100 may monitor the at least one resource element of the second set of resource elements for the downlink signal. Alternatively, the network node 104 may not transmit any signal on thesecond set of resource elements, in which case the downlink signal may be transmittedon another set of resource elements, on which the UE 100 may monitor for the downlink signal. The downlink signal may comprise at least one of: a data signal, a control signal, or another reference signal different from the downlink reference signal for which the downlink reference signal configuration is transmitted. For example, the downlink signal may comprise a downlink data channel (e.g., PDSCH) associated with the downlinkreference signal (e.g., PDSCH DMRS), in which case the downlink reference signal may beembedded within the downlink data channel to help the receiver (i.e., the UE 100) toestimate the channel conditions for demodulating and / or decoding the downlink datachannel. At 307, the UE 100 determines, per antenna port (e.g., for each antenna port)associated with the downlink reference signal (or per antenna port of the one or morefirst antenna ports), one or more channel estimates associated with the first set ofresource elements (received resource elements) in at least one of the frequency domainor the time domain. In other words, the one or more channel estimates are associatedwith the actual received resource elements of the downlink reference signal. The first setof resource elements refer to the resource elements on which the network node 104actually transmits the downlink reference signal. At 308, the UE 100 determines, using the channel predictor, per antenna port(e.g., for each antenna port) associated with the downlink reference signal (or perantenna port of the one or more second antenna ports), one or more predicted channelestimates associated with the second set of resource elements (non-received resourceelements) in at least one of the frequency domain or the time domain. In other words,the one or more predicted channel estimates are associated with the non-receivedresource elements of the downlink reference signal. The second set of resource elementsrefer to the resource elements on which the network node 104 does not transmit thedownlink reference signal (although the network node 104 may transmit anotherdownlink signal on the second set of resource elements, as explained above). The one or more predicted channel estimates may be determined based on the one or more channel estimates associated with the first set of resource elements (i.e.,the channel estimates obtained from the received resource elements may be used asinput for the channel predictor, together with the downlink reference signal configuration). At 309, the UE 100 stacks or combines the one or more channel estimates and the one or more predicted channel estimates, for example such that the combinationcovers the entire resource allocation of the downlink reference signal.At 310, the UE 100 demodulates and / or decodes the downlink signal (e.g., thedownlink data channel, such as PDSCH) based on the combination of the one or morechannel estimates and the one or more predicted channel estimates. FIG. 4 illustrates three examples of DL reference signal resource elementpatterns 410, 420, 430 (e.g., for DMRS type-1) associated with odd, even, and irregularallocation for a downlink data channel (e.g., PDSCH) allocation of four PRBs 401, 402,403, 404. In other words, FIG. 4 demonstrates how the reference signal resourceelements may be distributed across the frequency domain for different allocationpatterns. In the odd allocation pattern 410, the reference signal resource elements (i.e.,the first set of resource elements 441, 442, 443, 444, 445, 446, 453, 454, 455, 456, 457,458) are allocated to every odd-numbered PRB of the four PRBs allocated for the downlink data channel. This means that the reference signal resource elements (i.e., thefirst set of resource elements 441, 442, 443, 444, 445, 446, 453, 454, 455, 456, 457, 458)are placed in the first PRB 401 and in the third PRB 403. In this case, the second set ofresource elements 447, 448, 449, 450, 451, 452, 459, 460, 461, 462, 463, 464 (for whichthe channel predictor predicts the channel estimates) may be comprised in the even- numbered PRBs of the four PRBs allocated for the downlink data channel (i.e., in the second PRB 402 and in the fourth PRB 404). This pattern reduces overhead by using only the odd-numbered PRBs for the reference signal, allowing even-numbered PRBs to be used for any other DL signal (e.g., data signal, control signal, or another type of reference signal). In the even allocation pattern 420, the reference signal resource elements(i.e., the first set of resource elements 441, 442, 443, 444, 445, 446, 453, 454, 455, 456,457, 458) are allocated to every even-numbered PRB of the four PRBs allocated for thedownlink data channel. This means that the reference signal resource elements (i.e., thefirst set of resource elements 441, 442, 443, 444, 445, 446, 453, 454, 455, 456, 457, 458)are placed in the second PRB 402 and in the fourth PRB 404 of the downlink data channelallocation. In this case, the second set of resource elements 447, 448, 449, 450, 451, 452,459, 460, 461, 462, 463, 464 (for which the channel predictor predicts the channelestimates) may be comprised in the odd-numbered PRBs of the four PRBs allocated forthe downlink data channel (i.e., in the first PRB 401 and in the third PRB 403). Thispattern reduces overhead by using only the even-numbered PRBs for the reference signal, allowing odd-numbered PRBs to be used for any other DL signal (e.g., data signal, control signal, or another type of reference signal). In general, every n-th reference signal resource elements (i.e., the first set of resource elements) may be allocated to every n-th (n>1, for example n=2) PRB of the four PRBs allocated for the downlink data channel. For example, in case n=2, thereference signal resource elements (i.e., the first set of resource elements) are placed inthe second PRB 402 and in the fourth PRB 404 of the downlink data channel allocation. In this case, the second set of resource elements (for which the channel predictor predicts the channel estimates) may be comprised in the first PRB 401 and third PRB 403 of the four PRBs allocated for the downlink data channel. This pattern reduces overhead by using only the every n-th PRBs for the reference signal, allowing other PRBs to be used for any other DL signal (e.g., data signal, control signal, or another type of reference signal). The irregular allocation pattern 430 combines both the odd and evenallocations in an irregular manner. For example, the reference signal resource elements(i.e., the first set of resource elements) may be placed in the first PRB 401, the third PRB403 and the fourth PRB 404 (but not in the second PRB 402) allocated for the downlinkdata channel. In this case, the second set of resource elements (for which the channelpredictor predicts the channel estimates) may be comprised in the second PRB 402. Theirregular allocation can be tailored to specific channel conditions or networkrequirements, providing flexibility in optimizing the balance between resource overheadof the reference signal and data transmission efficiency.FIG. 5 illustrates a flow chart according to an example embodiment of amethod for indicating a channel prediction capability. The method of FIG. 5 may be performed by an apparatus 900 depicted in FIG.9. For example, the apparatus 900 may be, or comprise, or be comprised in, a user equipment (UE) 100, 102. Referring to FIG. 5, in block 501, the user equipment 100, 102 generates acapability indication (or a message comprising the capability indication), the capabilityindication indicating that the user equipment supports channel prediction associatedwith channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation. In block 502, the user equipment 100, 102 transmits the capability indicationto a network node 104.In block 503, the user equipment 100, 102 receives, from the network node 104, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor. The downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain. The transmission pattern comprises at least a first set of resource elementson which the network node 104 intends to transmit the downlink reference signal. Thetransmission pattern may further comprise a second set of resource elements on whichthe network node 104 does not intend to transmit the downlink reference signal.The downlink reference signal configuration may further comprise at least one of: an indication for operating the channel predictor in at least one of the frequency domain or the time domain, or at least one data set identifier from the one or more data set identifiers to be applied for the channel predictor. The first set of resource elements may be associated with one or more first antenna ports that are equivalent to or different than one or more second antenna ports associated with the second set of resource elements. The user equipment 100, 102 may receive, from the network node 104, thedownlink reference signal on the first set of resource elements. The user equipment 100, 102 may determine, per antenna port associated with the downlink reference signal, one or more channel estimates associated with the first set of resource elements in at least one of the frequency domain or the time domain. The user equipment 100, 102 may determine, using the channel predictor, per antenna port associated with the downlink reference signal, one or more predicted channel estimates associated with the second set of resource elements in at least one of the frequency domain or the time domain. The user equipment 100, 102 may demodulate a downlink data channelassociated with the downlink reference signal based on a combination of the one or more channel estimates and the one or more predicted channel estimates. The downlink reference signal is not received on the second set of resourceelements in at least one of the frequency domain or the time domain. The user equipment 100, 102 may receive a downlink signal from the network node on at least one resource element of the second set of resource elements in at least one of the frequency domain or the time domain. The downlink signal maycomprise at least one of: a data signal, a control signal, or another reference signaldifferent from the downlink reference signal for which the downlink reference signal configuration is received. The channel predictor may comprise a machine learning model 220 pre-trained based on a set of input data and a set of expected output data. The set of expected output data may comprise one or more reference signalchannel estimates in at least one of the frequency domain or the time domain per antenna port associated with the downlink reference signal. The set of input data may comprise at least one of: a physical resource blocklevel granularity associated with a resource element pattern type of the downlink reference signal in the frequency domain, a symbol position pattern of the downlinkreference signal in the time domain, a number of antenna ports associated with thedownlink reference signal, a sequence type of the downlink reference signal with one or more initialization seed values, or a downlink precoding technique of the downlink reference signal. The one or more data set identifiers may indicate at least one of: at least onephysical resource block density allocation of a reference signal resource element pattern type used for training the channel predictor, at least one physical resource block density allocation of a reference signal resource element pattern type that the channel predictoris capable of predicting, at least one reference signal symbol position pattern that thechannel predictor is capable of predicting in the time domain, one or more carrierfrequencies supported by the channel predictor, one or more numerology optionssupported by the channel predictor, one or more delay spread ranges supported by thechannel predictor, one or more Doppler frequency shift or spread values supported bythe channel predictor, one or more user equipment speed values supported by the channel predictor, one or more reference signal sequence initialization seed valuessupported by the channel predictor, a number of reference signal antenna portssupported by the channel predictor, one or more reference signal resource element typessupported by the channel predictor, one or more reference signal sequence typessupported by the channel predictor, a physical downlink shared channel allocationlength or range, in at least one of the frequency domain or the time domain, supportedby the channel predictor, one or more precoding types supported by the channelpredictor, or a precoding granularity, in at least one of the frequency domain or the timedomain, supported by the channel predictor. The downlink reference signal may comprise, for example, one of: a physicaldownlink shared channel demodulation reference signal, a channel state informationreference signal, or a phase-tracking reference signal.FIG. 6 illustrates a flow chart according to an example embodiment of amethod for indicating a channel prediction capability. The method of FIG. 6 may beperformed by an apparatus 1000 depicted in FIG.10. For example, the apparatus 1000 may be, or comprise, or be comprised in, a network node 104 of a radio access network. Referring to FIG.6, in block 601, the network node 104 receives, from a userequipment 100, 102, a capability indication indicating that the user equipment 100, 102supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation. In block 602, the network node 104 determines, based on the capability indication, a downlink reference signal configuration to be used for the channel predictor. The downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one of a frequency domain or a time domain. The transmission pattern comprises at least a first set of resource elementson which the network node 104 intends to transmit the downlink reference signal. Thetransmission pattern may further comprise a second set of resource elements on whichthe network node 104 does not intend to transmit the downlink reference signal.The downlink reference signal configuration may further comprise at least one of: an indication for operating the channel predictor in at least one of the frequency domain or the time domain, or at least one data set identifier from the one or more data set identifiers to be applied for the channel predictor. In block 603, the network node 104 transmits the downlink reference signalconfiguration to the user equipment 100, 102.The one or more data set identifiers may indicate at least one of: at least one physical resource block density allocation of a reference signal resource element pattern type used for training the channel predictor, at least one physical resource block density allocation of a reference signal resource element pattern type that the channel predictor is capable of predicting, at least one reference signal symbol position pattern that the channel predictor is capable of predicting in the time domain, one or more carrier frequencies supported by the channel predictor, one or more numerology options supported by the channel predictor, one or more delay spread ranges supported by the channel predictor, one or more Doppler frequency shift or spread values supported by the channel predictor, one or more user equipment speed values supported by the channel predictor, one or more reference signal sequence initialization seed values supported by the channel predictor, a number of reference signal antenna ports supported by the channel predictor, one or more reference signal resource element types supported by the channel predictor, one or more reference signal sequence types supported by the channel predictor, a physical downlink shared channel allocation length or range, in at least one of the frequency domain or the time domain, supported by the channel predictor, one or more precoding types supported by the channel predictor, or a precoding granularity, in at least one of the frequency domain or the time domain, supported by the channel predictor. The network node 104 may transmit the downlink reference signal to theuser equipment 100, 102 on the first set of resource elements, wherein the downlinkreference signal is not transmitted on the second set of resource elements in at least one of the frequency domain or the time domain. The network node 104 may transmit a downlink signal to the user equipment100, 102 on at least one resource element of the second set of resource elements in atleast one of the frequency domain or the time domain. The downlink signal may comprise at least one of: a data signal, a controlsignal, or another reference signal different from the downlink reference signal for whichthe downlink reference signal configuration is transmitted. The downlink reference signal may comprise one of: a physical downlinkshared channel demodulation reference signal, a channel state information referencesignal, or a phase-tracking reference signal.FIG. 7 illustrates a flow chart according to an example embodiment of amethod for improving data channel throughput. The method of FIG. 7 may be performedby an apparatus 900 depicted in FIG. 9. For example, the apparatus 900 may be, orcomprise, or be comprised in, a user equipment (UE) 100, 102, such as a smartphone orany other type of UE. This example embodiment may help to increase the capacity of the data channel such as PDSCH (i.e., increase the throughput of the data channel) by enabling transmission of data on some of the reference signal resource elements. Referring to FIG. 7, in block 701, the user equipment 100, 102 receives a downlink reference signal configuration from a network node 104, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain. For example, thedownlink reference signal configuration may be comprised in a radio resource controlmessage. The transmission pattern comprises a first set of resource elements on whichthe network node 104 intends to transmit the downlink reference signal, and a secondset of resource elements on which the network node 104 does not intend to transmit thedownlink reference signal. In block 702, the user equipment 100, 102 monitors the second set ofresource elements for a downlink data channel associated with the downlink reference signal. For example, the downlink reference signal may comprise a physical downlink shared channel demodulation reference signal, and the downlink data channelmay comprise a physical downlink shared channel.The monitoring may be performed based on the user equipment 100, 102 (or channel predictor) being configured to perform channel prediction associated with channel estimation on the second set of resource elements. In other words, if the UE 100,102 is configured with the channel prediction, the UE 100, 102 may assume the secondset of resource elements (not carrying the downlink reference signal information) to bemapped for the allocation of the downlink data channel (e.g., PDSCH allocation) on thescheduled time and / or frequency resources upon receiving the scheduling of thedownlink data channel from the network node 104.In block 703, based on the monitoring, the user equipment 100, 102 receives,from the network node 104, the downlink data channel on the second set of resourceelements. In other words, the UE 100 may receive the downlink data channel (e.g.,PDSCH) on the second set of resource elements for which the channel prediction may beconfigured, and that are configured in the configured time and / or frequency resourcesof the indicated allocation of the downlink data channel (e.g., PDSCH allocation).The first set of resource elements may be associated with one or more first antenna ports that are equivalent to or different than one or more second antenna ports that may be associated with the second set of resource elements. Prior to receiving the downlink reference signal configuration, the userequipment 100, 102 may generate a capability indication indicating that the userequipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channelestimation. The user equipment 100, 102 may transmit the capability indication to thenetwork node 104, and the downlink reference signal configuration may be received based on transmitting the capability indication. That is, the downlink reference signalconfiguration may be based on the capability indication.The user equipment 100, 102 may receive, from the network node 104, thedownlink reference signal on the first set of resource elements. The user equipment 100,102 may determine, per antenna port associated with the downlink reference signal, one or more channel estimates associated with the first set of resource elements in at least one of the frequency domain or the time domain. The user equipment 100, 102 may determine, using the channel predictor, per antenna port associated with the downlink reference signal, one or more predicted channel estimates associated with the secondset of resource elements in at least one of the frequency domain or the time domain. Theuser equipment 100, 102 may demodulate and / or decode the downlink data channelbased on a combination of the one or more channel estimates and the one or more predicted channel estimates. The channel predictor may comprise a machine learning model pre-trained based on a set of input data and a set of expected output data. The set of expected output data may comprise one or more reference signal channel estimates in at least one of the frequency domain or the time domain per antenna port associated with the downlinkreference signal. The set of input data may comprise at least one of: a physical resourceblock level granularity associated with a resource element pattern type of the downlink reference signal in the frequency domain, a symbol position pattern of the downlinkreference signal in the time domain, a number of antenna ports associated with thedownlink reference signal, a sequence type of the downlink reference signal with one ormore initialization seed values, or a downlink precoding technique of the downlink reference signal. The downlink reference signal configuration may further comprise at leastone of: an indication for operating the channel predictor in at least one of the frequencydomain or the time domain, or at least one data set identifier from the one or more dataset identifiers to be applied for the channel predictor. The one or more data set identifiers may indicate at least one physical resource block density allocation of a reference signal resource element pattern type used for training the channel predictor. Alternatively, the one or more data set identifiers may indicate at least onephysical resource block density allocation of a reference signal resource element pattern type that the channel predictor is capable of predicting. Alternatively, the one or more data set identifiers may indicate at least one reference signal symbol position pattern that the channel predictor is capable of predicting in the time domain. Alternatively, or additionally, the one or more data set identifiers indicate atleast one of: one or more carrier frequencies supported by the channel predictor, one ormore numerology options supported by the channel predictor, one or more delay spreadranges supported by the channel predictor, one or more Doppler frequency shift orspread values supported by the channel predictor, one or more user equipment speedvalues supported by the channel predictor, one or more reference signal sequenceinitialization seed values supported by the channel predictor, a number of referencesignal antenna ports supported by the channel predictor, one or more reference signalresource element types supported by the channel predictor, one or more reference signalsequence types supported by the channel predictor, a physical downlink shared channelallocation length or range, in at least one of the frequency domain or the time domain,supported by the channel predictor, one or more precoding types supported by thechannel predictor, or a precoding granularity, in at least one of the frequency domain orthe time domain, supported by the channel predictor. FIG. 8 illustrates a flow chart according to an example embodiment of amethod for improving data channel throughput. The method of FIG. 8 may be performedby an apparatus 1000 depicted in FIG. 10. For example, the apparatus 1000 may be, orcomprise, or be comprised in, a network node 104 of a radio access network.This example embodiment may help to increase the capacity of the datachannel such as PDSCH (i.e., increase the throughput of the data channel) by enabling transmission of data on some of the reference signal resource elements. Referring to FIG.8, in block 801, the network node 104 generates a downlink reference signal configuration, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one a frequency domain or a time domain. The transmission pattern comprises a first set of resource elements on whichthe network node 104 intends to transmit the downlink reference signal, and a secondset of resource elements on which the network node 104 does not intend to transmit the downlink reference signal.In block 802, the network node 104 transmits the downlink reference signalconfiguration to a user equipment 100.The network node 104 may transmit the downlink reference signal to the user equipment 100 on the first set of resource elements. The network node 104 may transmit, to the user equipment 100, on thesecond set of resource elements, a downlink data channel associated with the downlink reference signal. Prior to generating the downlink reference signal configuration, the networknode 104 may receive, from the user equipment 100, a capability indication indicatingthat the user equipment 100 supports channel prediction associated with channelestimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation. The downlink reference signal configuration may be generated based on the capability indication. The blocks, related functions, and information exchanges (messages)described above by means of FIG. 3 and FIGS. 5 to 8 are in no absolute chronologicalorder, and some of them may be performed simultaneously or in an order differing fromthe described 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. As used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements are joined by “and” or “or”, mean at least any one of the elements, or at least any two or more of the elements, or at least all the elements. FIG. 9 illustrates an example of an apparatus 900 comprising means forperforming one or more of the example embodiments (e.g., the method of FIG. 5 or FIG.7, or the functionalities of the UE 100 of FIG. 3) described above. For example, theapparatus 900 may be an apparatus such as, or comprising, or comprised in, a userequipment (UE) 100, 102. The user equipment may also be called a wirelesscommunication device, a subscriber unit, a mobile station, a remote terminal, an access terminal, a user terminal, a terminal device, or a user device. The apparatus 900 may comprise a circuitry or a chipset applicable for realizing one or more of the example embodiments described above. For example, the apparatus 900 may comprise at least one processor 910. The at least one processor 910interprets instructions (e.g., computer program instructions) and processes data. The atleast one processor 910 may comprise one or more programmable processors. The at least one processor 910 may comprise programmable hardware with embedded firmware and may, alternatively or additionally, comprise one or more application- specific integrated circuits (ASICs). The at least one processor 910 is coupled to at least one memory 920. The at least one processor is configured to read and write data to and from the at least one memory 920. The at least one memory 920 may comprise one or more memory units. The memory units may be volatile or non-volatile. It is to be noted that there may be one or more units of non-volatile memory and one or more units of volatile memory or, alternatively, one or more units of non-volatile memory, or, alternatively, one or more units of volatile memory. Volatile memory may be for example random-access memory (RAM), dynamic random-access memory (DRAM) or synchronous dynamic random- access memory (SDRAM). Non-volatile memory may be for example read-only memory (ROM), programmable read-only memory (PROM), electronically erasable programmable read-only memory (EEPROM), flash memory, optical storage or magnetic storage. In general, memories may be referred to as non-transitory computer readable media. 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). The at least one memory 920 stores computer readable instructions that are executed by the at least one processor 910 to perform one or more of the example embodiments described above. For example, non-volatile memory stores the computer readable instructions, and the at least one processor 910 executes the instructions using volatile memory for temporary storage of data and / or instructions. The computer readable instructions may refer to computer program code. The computer readable instructions may have been pre-stored to the at least one memory 920 or, alternatively or additionally, they may be received, by the apparatus, via an electromagnetic carrier signal and / or may be copied from a physical entity such as a computer program product. Execution of the computer readable instructions by the at least one processor 910 causes the apparatus 900 to perform one or more of the example embodiments described above. That is, the at least one processor and the at least one memory storing the instructions may provide the means for providing or causing the performance of any of the methods and / or blocks described above. In the context of this document, a “memory” or “computer-readable media” or “computer-readable medium” may be any non-transitory media or medium or means that can contain, store, communicate, propagate or transport the instructions for use by or in connection with an instruction execution system, apparatus, or device, such as a computer. 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). The apparatus 900 may further comprise, or be connected to, an input unit 930. The input unit 930 may comprise one or more interfaces for receiving input. The one or more interfaces may comprise for example one or more temperature, motion and / or orientation sensors, one or more cameras, one or more accelerometers, one or more microphones, one or more buttons and / or one or more touch detection units. Further, the input unit 930 may comprise an interface to which external devices may connect to. The apparatus 900 may also comprise an output unit 940. The output unit may comprise or be connected to one or more displays capable of rendering visual content, such as a light emitting diode (LED) display, a liquid crystal display (LCD) and / or a liquid crystal on silicon (LCoS) display. The output unit 940 may further comprise one or more audio outputs. The one or more audio outputs may be for example loudspeakers. The apparatus 900 further comprises a connectivity unit 950. The connectivity unit 950 enables wireless connectivity to one or more external devices. The connectivity unit 950 comprises at least one transmitter and at least one receiver that may be integrated to the apparatus 900 or that the apparatus 900 may be connected to. The at least one transmitter comprises at least one transmission antenna, and the at least one receiver comprises at least one receiving antenna. The connectivity unit 950 may comprise an integrated circuit or a set of integrated circuits that provide the wireless communication capability for the apparatus 900. Alternatively, the wireless connectivity may be a hardwired application-specific integrated circuit (ASIC). The connectivity unit950 may also provide means for performing at least some of the blocks or functions of one or more example embodiments described above. The connectivity unit 950 may comprise one or more components, such as: power amplifier, digital front end (DFE), analog-to-digital converter (ADC), digital-to-analog converter (DAC), frequency converter, (de)modulator, and / or encoder / decoder circuitries, controlled by the corresponding controlling units. It is to be noted that the apparatus 900 may further comprise various components not illustrated in FIG. 9. The various components may be hardware components and / or software components. FIG. 10 illustrates an example of an apparatus 1000 comprising means forperforming one or more of the example embodiments (e.g., the method of FIG. 6 or FIG.8, or the functionalities of the network node 104 of FIG.3) described above. For example,the apparatus 1000 may be an apparatus such as, or comprising, or comprised in, anetwork node 104 of a radio access network.The apparatus 1000 may comprise, for example, a circuitry or a chipset applicable for realizing one or more of the example embodiments described above. The apparatus 1000 may be an electronic device comprising one or more electroniccircuitries. The apparatus 1000 may comprise a communication control circuitry 1010such as at least one processor, and at least one memory 1020 storing instructions 1022which, when executed by the at least one processor, cause the apparatus 1000 to carryout one or more of the example embodiments described above. Such instructions 1022may, for example, include computer program code (software). The at least one processor and the at least one memory storing the instructions may provide the means for providing or causing the performance of any of the methods and / or blocks described above. The processor is coupled to the memory 1020. The processor is configured to read and write data to and from the memory 1020. The memory 1020 may comprise one or more memory units. The memory units may be volatile or non-volatile. It is to be noted that there may be one or more units of non-volatile memory and one or more units of volatile memory or, alternatively, one or more units of non-volatile memory, or, alternatively, one or more units of volatile memory. Volatile memory may be for example random-access memory (RAM), dynamic random-access memory (DRAM) or synchronous dynamic random-access memory (SDRAM). Non-volatile memory may be for example read-only memory (ROM), programmable read-only memory (PROM), electronically erasable programmable read-only memory (EEPROM), flash memory, optical storage or magnetic storage. In general, memories may be referred to as non- transitory computer readable media. The term “non-transitory,” as used herein, is alimitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation ondata storage persistency (e.g., RAM vs. ROM). The memory 1020 stores computer readable instructions that are executed by the processor. For example, non-volatile memory stores the computer readable instructions, and the processor executes the instructions using volatile memory for temporary storage of data and / or instructions. The computer readable instructions may have been pre-stored to the memory 1020 or, alternatively or additionally, they may be received, by the apparatus, via an electromagnetic carrier signal and / or may be copied from a physical entity such as a computer program product. Execution of the computer readable instructions causes the apparatus 1000 to perform one or more of the functionalities described above. The memory 1020 may be implemented using any suitable data storagetechnology, such as semiconductor-based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, fixed memory and / or removable memory. The memory may comprise a configuration database for storing configuration data, such as a current neighbour cell list, and, in some exampleembodiments, structures of frames used in the detected neighbour cells.The apparatus 1000 may further comprise or be connected to acommunication interface 1030, such as a radio unit, comprising hardware and / orsoftware for realizing communication connectivity with one or more wireless communication devices according to one or more communication protocols. The communication interface 1030 comprises at least one transmitter (Tx) and at least one receiver (Rx) that may be integrated to the apparatus 1000 or that the apparatus 1000 may be connected to. The communication interface 1030 may provide means forperforming some of the blocks and / or functions (e.g., transmitting and receiving) for oneor more example embodiments described above. The communication interface 1030 may comprise one or more components, such as: power amplifier, digital front end (DFE), analog-to-digital converter (ADC), digital-to-analog converter (DAC), frequency converter, (de)modulator, and / or encoder / decoder circuitries, controlled by the corresponding controlling units. The communication interface 1030 provides the apparatus with radio communication capabilities to communicate in the wireless communication network. The communication interface may, for example, provide a radio interface to one or moreUEs 100, 102. The apparatus 1000 may further comprise or be connected to anotherinterface towards a core network 110, such as the network coordinator apparatus orAMF, and / or to other access nodes of the wireless communication network.The apparatus 1000 may further comprise a scheduler 1040 that is configured to allocate radio resources. The scheduler 1040 may be configured along with the communication control circuitry 1010 or it may be separately configured. It is to be noted that the apparatus 1000 may further comprise various components not illustrated in FIG. 10. The various components may be hardware components and / or software components. 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, 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 mobile phone, to perform various functions); and c) hardware circuit(s) and / or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (for example firmware) for operation, but the software may not be present when it is not needed for operation. This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device. The techniques and methods described herein may be implemented by various means. For example, these techniques may be implemented in hardware (one or more devices), firmware (one or more devices), software (one or more modules), or combinations thereof. For a hardware implementation, the apparatus(es) of example embodiments may be implemented within one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), graphics processing units (GPUs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof. For firmware or software, the implementation can be carried out through modules of at least one chipset (for example procedures, functions, and so on) that perform the functions described herein. The software codes may be stored in a memory unit and executed by processors. The memory unit may be implemented within the processor or externally to the processor. In the latter case, it can be communicatively coupled to the processor via various means, as is known in the art. Additionally, the components of the systems described herein may be rearranged and / or complemented by additional components in order to facilitate the achievements of the various aspects, etc., described with regard thereto, and they are not limited to the precise configurations set forth in the given figures, as will be appreciated by one skilled in the art. It will be obvious to a person skilled in the art that, as technology advances,the inventive concept may be implemented in various ways within the scope of theclaims. The embodiments are not limited to the example embodiments described above, but may vary within the scope of the claims. Therefore, all words and expressions should be interpreted broadly, and they are intended to illustrate, not to restrict, the embodiments.

Claims

1. CLAIMS 1. A user equipment comprising at least one processor, and at least one memory storing instructions that, when executed by the at least one processor, cause the user equipment at least to: receive a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitor the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receive, from the network node, the downlink data channel on the second set of resource elements based on the monitoring.

2. The user equipment of claim 1, wherein the monitoring is performed based on the user equipment being configured to perform channel prediction associated with channel estimation on the second set of resource elements.

3. The user equipment of any preceding claim, further being caused to: generate a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; and transmit the capability indication to the network node, wherein the downlink reference signal configuration is based on the capability indication.

4. The user equipment of claim 3, further being caused to: receive, from the network node, the downlink reference signal on the first set of resource elements;determine, per antenna port associated with the downlink reference signal, one or more channel estimates associated with the first set of resource elements in at least one of the frequency domain or the time domain; determine, using the channel predictor, per antenna port associated with the downlink reference signal, one or more predicted channel estimates associated with the second set of resource elements in at least one of the frequency domain or the time domain; and demodulate the downlink data channel based on a combination of the one or more channel estimates and the one or more predicted channel estimates.

5. The user equipment of any of claims 3 to 4, wherein the channel predictorcomprises a machine learning model pre-trained based on a set of input data and a set of expected output data, wherein the set of expected output data comprises one or more reference signal channel estimates in at least one of the frequency domain or the time domain per antenna port associated with the downlink reference signal, wherein the set of input data comprises at least one of: a physical resource block level granularity associated with a resource element pattern type of the downlink reference signal in the frequency domain, a symbol position pattern of the downlink reference signal in the time domain, a number of antenna ports associated with the downlink reference signal, a sequence type of the downlink reference signal with one or more initialization seed values, or a downlink precoding technique of the downlink reference signal.

6. The user equipment of any of claims 3 to 5, wherein the downlink reference signal configuration further comprises at least one of: an indication for operating the channel predictor in at least one of the frequency domain or the time domain, or at least one data set identifier from the one or more data set identifiers to be applied for the channel predictor.

7. The user equipment of any of claims 3 to 6, wherein the one or more data set identifiers indicate at least one physical resource block density allocation of a reference signal resource element pattern type used for training the channel predictor.

8. The user equipment of any of claims 3 to 6, wherein the one or more data set identifiers indicate at least one physical resource block density allocation of a reference signal resource element pattern type that the channel predictor is capable of predicting.

9. The user equipment of any of claims 3 to 8, wherein the one or more data set identifiers indicate at least one of: one or more carrier frequencies supported by the channel predictor, one or more numerology options supported by the channel predictor, one or more delay spread ranges supported by the channel predictor, one or more Doppler frequency shift or spread values supported by the channel predictor, one or more user equipment speed values supported by the channel predictor, one or more reference signal sequence initialization seed values supported by the channel predictor, a number of reference signal antenna ports supported by the channel predictor, one or more reference signal resource element types supported by the channel predictor, one or more reference signal sequence types supported by the channel predictor, a physical downlink shared channel allocation length or range, in at least one of the frequency domain or the time domain, supported by the channel predictor, one or more precoding types supported by the channel predictor, or a precoding granularity, in at least one of the frequency domain or the time domain, supported by the channel predictor.

10. The user equipment of any preceding claim,wherein the downlink reference signal is a physical downlink shared channel demodulation reference signal, wherein the downlink data channel is a physical downlink shared channel.

11. A network node comprising at least one processor, and at least one memory storing instructions that, when executed by the at least one processor, cause the network node at least to: generate a downlink reference signal configuration, wherein the downlink reference signal configuration comprises a transmission pattern for a downlink reference signal in at least one a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; transmit the downlink reference signal configuration to a user equipment; and transmit, to the user equipment, on the second set of resource elements, a downlink data channel associated with the downlink reference signal.

12. A method performed by a user equipment, the method comprising: receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitoring the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring.

13. The method of claim 12, wherein the monitoring is performed based on the user equipment being configured to perform channel prediction associated with channel estimation on the second set of resource elements.

14. The method of claim 12 or 13, further comprising: generating a capability indication indicating that the user equipment supports channel prediction associated with channel estimation, wherein the capability indication comprises one or more data set identifiers indicating a reference signal configuration used for training of a channel predictor associated with the channel estimation; and transmit the capability indication to the network node, wherein the downlink reference signal configuration is based on the capability indication.

15. A non-transitory computer readable medium comprising program instructions which, when executed by a user equipment, cause the user equipment to perform at least the following: receiving a downlink reference signal configuration from a network node, wherein the downlink reference signal configuration comprises a transmission pattern of a downlink reference signal in at least one of a frequency domain or a time domain, the transmission pattern comprising a first set of resource elements on which the network node intends to transmit the downlink reference signal, and a second set of resource elements on which the network node does not intend to transmit the downlink reference signal; monitoring the second set of resource elements for a downlink data channel associated with the downlink reference signal; and receiving, from the network node, the downlink data channel on the second set of resource elements based on the monitoring.

Images