Cold start channel prediction

Abstract

An apparatus configured to determine to perform channel prediction and reporting during an observation window, determine, during the observation window, at least one of at least one channel prediction based at least in part on one or more observations of a channel made during the observation window or at least one observation of the channel made during the observation window, and transmit the determined at least one of the at least one channel prediction or the at least one channel observation during the observation window.

Description

Technical Field

[0001] The example and non-limiting embodiments generally relate to channel prediction, and more specifically, to channel prediction in the context of a long observation window. Background Technology

[0002] In channel prediction, it is known that predictions for time slots are reported during the prediction window after the observation window is completed. Summary of the Invention

[0003] The following overview is intended to be illustrative only. The summary is not intended to limit the scope of the claims.

[0004] According to one aspect, an apparatus includes: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: determine to perform channel prediction and reporting during an observation window; determine at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window; and transmit at least one of the at least one channel prediction or at least one observation of the channel determined during the observation window.

[0005] According to one aspect, a method includes: using a user equipment to determine performing channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window; and transmitting at least one of the at least one channel prediction or at least one observation of the channel during the observation window.

[0006] According to one aspect, an apparatus includes a component for causing the apparatus to perform at least: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of a channel made during the observation window; and transmitting at least one of the at least one channel prediction or at least one channel observation of a channel during the observation window.

[0007] According to one aspect, a computer-readable medium includes instructions stored thereon for performing at least the following operations: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window; and causing at least one of the at least one determined in the at least one channel prediction or at least one channel observation of the channel to be transmitted during the observation window.

[0008] According to one aspect, an apparatus includes: at least one processor; and at least one memory, the memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: send an instruction to a user equipment to perform channel prediction and reporting during an observation window; receive, during the observation window, at least one of: at least one channel prediction made during the observation window, or at least one observation of a channel made during the observation window; and transmit downlink data based at least in part on at least one of the at least one channel prediction or at least one observation of a channel.

[0009] According to one aspect, a method includes: using a network node to send an instruction to a user equipment to perform channel prediction and reporting during an observation window; receiving, during the observation window, at least one of the following: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and transmitting downlink data based at least in part on at least one of the at least one channel prediction or at least one observation of the channel.

[0010] According to one aspect, an apparatus includes a component for causing the apparatus to at least perform: sending an instruction to a user equipment to perform channel prediction and reporting during an observation window; receiving, during the observation window, at least one of the following: at least one channel prediction made during the observation window, or at least one observation of a channel made during the observation window; and transmitting downlink data based at least in part on at least one of the at least one channel prediction or at least one observation of a channel.

[0011] According to one aspect, a computer-readable medium includes instructions stored thereon for performing at least the following: causing a network node to send an instruction to a user equipment to perform channel prediction and reporting during an observation window; causing at least one of the following to be received during the observation window: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and causing downlink data to be transmitted based at least in part on at least one of the at least one channel prediction or at least one observation of the channel.

[0012] The independent claims provide the subject matter for several aspects. Additional aspects are defined in the dependent claims. Attached Figure Description

[0013] The foregoing aspects and other features are explained in the following description in conjunction with the accompanying drawings, wherein: Figure 1 This is a block diagram of one possible and non-limiting example system in which exemplary embodiments can be practiced; Figure 2 This is a diagram illustrating the features as described herein; Figure 3 This is a diagram illustrating the features as described herein; Figure 4 This is a diagram illustrating the features as described herein; Figure 5 This is a flowchart illustrating the steps described herein; Figure 6 This is a flowchart illustrating the steps described herein; Figure 7 This is a flowchart illustrating the steps described herein; Figure 8 This is a flowchart illustrating the steps described herein; and Figure 9 This is a flowchart illustrating the steps described herein. Detailed Implementation

[0014] The following abbreviations, which can be found in the instruction manual and / or accompanying drawings, are defined as follows: 3GPP Third Generation Partnership Project 5G fifth generation 5GC 5G Core Network AI (Artificial Intelligence) AMF access and mobility management functions CE control elements CN Core Network CQI Channel Quality Indicator cRAN Cloud Radio Access Network CSI Channel Status Information CU Central Unit DCI downlink control indicator DL downlink DNN deep neural network DU Distributed Unit eNB (or eNodeB) evolved Node B (e.g., LTE base station) EN-DCE-UTRA-NR Dual Connection The en-gNB or En-gNB provides the UE with NR user plane and control plane protocol termination and acts as a secondary node in the EN-DC. E-UTRA evolved universal terrestrial radio access, i.e., LTE radio access technology Generative Adversarial Network (GAN) gNB (or gNodeB) is a base station used for 5G / NR, that is, a node that provides NR user plane and control plane protocol termination to the UE and connects to the 5GC via the NG interface. I / F interface IP Internet Protocol KPIs (Key Performance Indicators) L1 First Floor LTE Long Term Evolution MAC Media Access Control MCS modulation and coding scheme MIMO (Multiple Input Multiple Output) ML machine learning MME Mobility Management Entity MSE Mean Square Error MSOML Split Arranger ng or NG next generation ng-eNB or NG-eNB next-generation eNB NN Neural Network NR New Radio N / W or NW network O-RAN Open Radio Access Network PDCCH Physical Downlink Control Channel PDCP Packet Data Convergence Protocol PHY physical layer RAN Radio Access Network RE Resource Elements RF radio frequency RLC Radio Link Control RRC Radio Resource Control RRH Remote Radio Header RS reference signal RU radio unit Rx receiver SDAP Service Data Adaptation Protocol SGCS squared generalized cosine similarity SGW Service Gateway SMF Session Management Function Tx transmitter UE (User Equipment) (e.g., wireless, typically mobile devices) UPF User Plane Functions ZOH holds at zero order Go to Figure 1 The figure illustrates a block diagram of one possible and non-limiting example in which practical examples can be implemented. It shows a user equipment (UE) 110, a radio access network (RAN) node 170, and one or more network elements 190. Figure 1 In the example, User Equipment (UE) 110 wirelessly communicates with Wireless Network 100. The UE is a wireless device that can access Wireless Network 100. UE 110 includes one or more processors 120, one or more memories 125, and one or more transceivers 130 interconnected via one or more buses 127. Each of the one or more transceivers 130 includes a receiver (Rx) 132 and a transmitter (Tx) 133. The one or more buses 127 may be address, data, or control buses and may include any interconnection mechanism, such as a series of lines on a motherboard or integrated circuit, fiber optic cables, or other optical communication devices. "Circuit" may include dedicated hardware or hardware associated with software that can be executed thereon. The one or more transceivers 130 are connected to one or more antennas 128. The one or more memories 125 include computer-readable code 123. UE 110 includes a module 140, which includes one or both of portions 140-1 and / or 140-2, which may be implemented in various ways. Module 140 may be implemented in hardware as module 140-1, such as as part of one or more processors 120. Module 140-1 may also be implemented as an integrated circuit or by other hardware such as a programmable gate array. In another example, module 140 may be implemented as module 140-2, which is implemented as computer-readable code 123 and executed by one or more processors 120. For example, one or more memories 125 and computer-readable code 123 may be configured to cause user equipment 110 to perform one or more of the operations described herein via one or more processors 120. UE 110 communicates with RAN node 170 via radio link 111.

[0015] In this example, RAN node 170 is a base station that provides access to wireless network 100 by wireless devices such as UE 110. RAN node 170 can be, for example, a base station for 5G (also known as New Radio (NR)). In 5G, RAN node 170 can be an NG-RAN node, which is defined as a gNB or ng-eNB. A gNB is a node that provides NR user plane and control plane protocol termination to the UE and is connected to the 5GC (e.g., network element 190) via an NG interface. An ng-eNB is a node that provides E-UTRA user plane and control plane protocol termination to the UE and is connected to the 5GC via an NG interface. An NG-RAN node can include multiple gNBs, and can also include a central unit (CU) (gNB-CU) 196 and a distributed unit (DU) (gNB-DU), where DU 195 is shown. Note that a DU can include or be coupled to and control a radio unit (RU). A gNB-CU is a logical node that hosts the RRC, SDAP, and PDCP protocols of the gNB or controls the RRC and PDCP protocols of an en-gNB that controls the operation of one or more gNB-DUs. The gNB-CU terminates the F1 interface connected to the gNB-DU. The F1 interface is shown as reference numeral 198, although reference numeral 198 also shows the link between remote elements of RAN node 170 and centralized elements of RAN node 170 (such as between gNB-CU 196 and gNB-DU 195). The gNB-DU is a logical node that hosts the RLC, MAC, and PHY layers of the gNB or en-gNB, and its operation is partially controlled by the gNB-CU. One gNB-CU supports one or more cells. A cell is supported by only one gNB-DU. The gNB-DU terminates the F1 interface 198 connected to the gNB-CU. Note that DU 195 is considered to include transceiver 160, for example, as part of the RU, but some examples may have transceiver 160 as part of a separate RU, for example, under the control of DU 195 and connected to DU 195. RAN node 170 can also be an eNB (evolved NodeB) base station for LTE (Long Term Evolution), or any other suitable base station, access point, access node, or node.

[0016] RAN node 170 includes one or more processors 152, one or more memories 155, one or more network interfaces (N / WI / F) 161, and one or more transceivers 160 interconnected via one or more buses 157. Each of the one or more transceivers 160 includes a receiver (Rx) 162 and a transmitter (Tx) 163. The one or more transceivers 160 are connected to one or more antennas 158. The one or more memories 155 include computer-readable code 153. CU 196 may include one or more processors 152, memories 155, and network interfaces 161. Note that DU 195 may also include its own memories / multiple memories and one or more processors and / or other hardware, but these are not shown.

[0017] RAN node 170 includes module 150, which includes one or both of portions 150-1 and / or 150-2, and can be implemented in various ways. Module 150 can be implemented in hardware as module 150-1, such as as a portion of one or more processors 152. Module 150-1 can also be implemented as an integrated circuit or by other hardware such as a programmable gate array. In another example, module 150 can be implemented as module 150-2, which is implemented as computer-readable code 153 and executed by one or more processors 152. For example, one or more memories 155 and computer-readable code 153 are configured to cause RAN node 170 to perform one or more operations as described herein via one or more processors 152. Note that the functionality of module 150 can be distributed, such as distributed between DU 195 and CU 196, or implemented only in DU 195.

[0018] One or more network interfaces 161 communicate via a network, such as via links 176 and 131. Two or more gNBs 170 can communicate using, for example, link 176. Link 176 can be wired, wireless, or both, and can implement, for example, an Xn interface for 5G, an X2 interface for LTE, or other suitable interfaces for other standards.

[0019] One or more buses 157 may be address, data, or control buses and may include any interconnection mechanism, such as a series of lines on a motherboard or integrated circuit, optical fiber or other optical communication equipment, wireless channels, etc. For example, one or more transceivers 160 may be implemented as a Remote Radio Header (RRH) 195 for LTE or a Distributed Unit (DU) 195 for a gNB implementation of 5G, wherein other elements of the RAN node 170 may be physically located in a different location from the RRH / DU, and one or more buses 157 may be partially implemented as, for example, fiber optic cables or other suitable network connections to connect other elements of the RAN node 170 (e.g., Central Unit (CU), gNB-CU) to the RRH / DU 195. Reference numeral 198 also indicates those suitable network links(s).

[0020] Note that the description in this document indicates that a "cell" performs a function; however, it should be clear that the equipment forming the cell will perform the function. Cells constitute part of a base station. That is, each base station can have multiple cells. For example, for a single carrier frequency and associated bandwidth, there can be three cells, each covering one-third of a 360-degree area, such that the coverage area of ​​a single base station is approximately elliptical or circular. Furthermore, each cell can correspond to a single carrier, and a base station can use multiple carriers. Therefore, if each carrier has three 120-degree cells and two carriers, the base station has a total of six cells.

[0021] Wireless network 100 may include one or more network elements 190, which may include core network functions and provide connectivity to another network (such as a telephone network and / or a data communication network (e.g., the Internet)) via one or more links 181. Such core network functions for 5G may include (one or more) Access and Mobility Management Functions (AMF) and / or (one or more) User Plane Functions (UPF) and / or (one or more) Session Management Functions (SMF). Such core network functions for LTE may include MME (Mobility Management Entity) / SGW (Serving Gateway) functions. These are merely illustrative functions that may be supported by (one or more) network elements 190, and it should be noted that both 5G and LTE functions may be supported. RAN node 170 is coupled to network element 190 via link 131. Link 131 may be implemented as, for example, an NG interface for 5G, or an S1 interface for LTE, or other suitable interfaces for other standards. Network element 190 includes one or more processors 175, one or more memories 171, and one or more network interfaces (N / WI / F) 180 interconnected via one or more buses 185. One or more memories 171 include computer-readable code 173. The one or more memories 171 and computer-readable code 173 are configured to cause network element 190 to perform one or more operations via one or more processors 175.

[0022] Wireless network 100 can implement network virtualization, which is the process of combining hardware and software network resources and network functions into a single software-based managed entity (virtual network). Network virtualization involves platform virtualization, often combined with resource virtualization. Network virtualization is categorized as external (combining many networks or parts of networks into virtual units) or internal (providing network-like functionality to a software container on a single system). For example, a network can be deployed in a remote cloud, where virtualized network functions (VNFs) run on, for example, data center servers. For instance, network core functions and / or (one or more) radio access networks (e.g., CloudRAN, O-RAN, edge cloud) can be virtualized. Note that the virtualized entity resulting from network virtualization is still implemented to some extent using hardware such as processors 152 or 175 and memories 155 and 171, and such virtualized entities also produce technical effects.

[0023] It should also be noted that the operation of the example embodiments of this disclosure may be performed by multiple cooperating devices (e.g., cRAN).

[0024] Computer-readable storage devices 125, 155, and 171 can be of any type suitable for the local technical environment and can be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic storage devices and systems, optical storage devices and systems, fixed storage, and removable storage. Computer-readable storage devices 125, 155, and 171 can be components for performing storage functions. As a non-limiting example, processors 120, 152, and 175 can be of any type suitable for the local technical environment and can include one or more of a general-purpose computer, a special-purpose computer, a microprocessor, a digital signal processor (DSP), and a processor based on a multi-core processor architecture. Processors 120, 152, and 175 can be components for performing functions such as controlling UE 110, RAN node 170, and other functions as described herein.

[0025] Typically, various example embodiments of user equipment 110 may include, but are not limited to, cellular phones (such as smartphones), tablet computers, personal digital assistants (PDAs) with wireless communication capabilities, portable computers with wireless communication capabilities, image capture devices (such as digital cameras) with wireless communication capabilities, gaming devices with wireless communication capabilities, music storage and playback devices with wireless communication capabilities, internet devices that allow wireless internet access and browsing, tablet computers with wireless communication capabilities, and portable units or terminals that include a combination of these functions.

[0026] Therefore, a suitable but non-limiting technical context has been introduced for practicing exemplary embodiments of this disclosure, and the exemplary embodiments will now be described in more detail.

[0027] The features described herein can broadly relate to channel prediction. In Release 18 (Rel. 18), non-AI / ML-based channel prediction is defined as an observation window over K time slots, which may span a duration of 24 ms. Based on observations and / or measurements made during the observation window, the channel conditions during the prediction window are predicted. In this disclosure, the terms "observation" and "measurement" are used interchangeably to refer to information determined by the UE based on the Channel State Information (CSI) Reference Signal (RS) received during the observation window, which is used to estimate the radio channel.

[0028] It is anticipated that the primary functionality of non-AI / ML-based channel prediction will be reused for AI / ML-based channel prediction. In the AI / ML-based channel prediction 3GPP Release 19 (Rel. 19) research project, a 10×5ms observation window (i.e., the number of CSI-RS signals multiplied by the number of time slots between signals) has been used, equaling a duration of 50ms. For channel prediction on the UE side, the UE then reports the predicted CSIs occurring within its relevant prediction window for N4 time slots. Since the prediction window follows the observation window, there will always be a delay of at least, for example, 24 to 50ms before the UE can trigger the reporting of the predicted CSIs. The gNB must wait for another N4 = 1 to 4 time slots to receive the predicted CSIs before it can begin its downlink transmission. This high latency is a drawback, especially in applications requiring low latency or high data rates.

[0029] The technical effect of the exemplary embodiments of this disclosure can be to reduce large startup delays during channel prediction modes, whether based on non-AI / ML or AI / ML.

[0030] The exemplary embodiments of this disclosure can be applied to the context of artificial intelligence (AI) and / or machine learning (ML). One or more AI / ML models may reside in each of the UE and the base station (e.g., split across multiple nodes, or as a separate model at each node). Alternatively, the AI / ML model may reside in either the UE or the base station. This disclosure focuses on channel prediction on the UE side, where the ML model is implemented at the UE device, but this is not limiting.

[0031] An example of an AI / ML model is a neural network. Other examples can include deep neural networks (DNNs), generative adversarial networks (GANs), or ML segmentation orchestrators (MSOs). A neural network (NN) is a computational graph consisting of two or more layers of computation. Each layer can consist of one or more units, each of which can perform basic computations. Units can be connected to one or more other units, and this connection can have weights associated with it. Weights can be used to scale signals through the associated connections. Weights can be learnable parameters, i.e., values ​​that can be learned from training data. Other learnable parameters can exist, such as the parameters of batch normalization layers.

[0032] Two of the most widely used architectures for neural networks are feedforward architectures and recursive architectures. Feedforward neural networks do not include feedback loops; each layer takes input from one or more previous layers and provides an output that is used as input to one or more subsequent layers. Units within a layer take input from one or more units in one or more previous layers and provide output to one or more units in one or more subsequent layers.

[0033] The initial layer, which is closest to the input data, extracts low-level semantic features from the received data, while the intermediate and final layers extract higher-level features. Following the feature extraction layers, one or more layers can perform specific tasks, such as classification, semantic segmentation, object detection, denoising, style transfer, and super-resolution. In recurrent neural networks, feedback loops exist, making the network stateful; that is, it can remember or retain information or states.

[0034] Neural networks and other machine learning tools may be able to learn properties from input data in a supervised or unsupervised manner. This learning can be the result of training algorithms or meta-neural networks that provide training signals.

[0035] Now for reference Figure 2 This illustrates an example of a non-AI / ML-based Channel State Information (CSI) Reference Signal (RS) configuration framework for channel prediction. The configuration begins with a dark gray cross (220) indicating... A non-periodic CSI RS (250) trigger (240), which has an equal number of one or two time slots (i.e., The relative delay of m(260) is ms or 2ms. The CSI report indicated by the black cross (210) is triggered by passing through or using the Type II Doppler codebook (e.g., W at N4 time instances). CSI =Type II codebook report) begins in the prediction The UE reports the predicted CSI at point (280) (indicated by the light gray cross (230)). The UE sends the first CSI report at time instance n+δ (270) after the CSI report trigger (210) and / or after the last CSI RS in the observation window. n and δ are configured by the gNB. n is the reference point relative to the first possible CSI report from the last CSI RS in the observation window (250). δ defines the relative delay with n and is used to ensure that the time instances for N4 CSI reports are consistent with the DL precoding time (fall together). Depending on the configuration, the UE reports (280) equals the type II codebook CSI predicted by {1,2,4,8}. The point to be predicted (280) can have d between them. The delay (290) of {1, m} time slots, where m can be equivalent to 1ms or 2ms. {1,2} time slots.

[0036] Figure 2 One drawback of the Rel 18-based non-AI / ML channel prediction CSI RS framework is the delay associated with the observation window (250), which may have... The length is ms = 24ms (i.e., K×m). For AI / ML-based channel prediction in Rel 19, observation windows of even longer, such as 50ms or even longer, have been evaluated. Such a long observation window means that there will be a long delay of 24ms or longer from the start of sending aperiodic CSI RS (240) until the reporting trigger (210) of the CSI (230) used for prediction. The time instance to the UE reporting the predicted CSI, and the duration of further gNB using the predicted CSI report for DL ​​precoding of user data, is even longer.

[0037] Given that the average latency for NR can be in the range of a few milliseconds, such additional latency of approximately 25 to 50 milliseconds for channel prediction is a drawback. This is particularly problematic for Internet Protocol (IP) services, potentially having an additional impact on the maximum data rate, which decreases as the latency value increases. The technical effect of the example embodiments of this disclosure can be to reduce the additional latency due to channel prediction in the range of 20 to 50 milliseconds.

[0038] In the example implementation, a low-latency cold start mode can be implemented for channel prediction, which can begin at the start of the observation window. In other words, the UE can make one or more channel state predictions before the CSI report is triggered, rather than making all channel state predictions after the CSI report is triggered. Figure 3 An example of a cold-start prediction mode configuration is shown, where the UE can, after the start of a triggered AP-CSI-RS (325), report a first short-term CSI prediction (320) for a point to be predicted (315) or a point during the observation window (330) based on one or more aperiodic CSI-RS burst measurements (310) spaced at m time slots (335), either during or at the start of the observation window (330). This means that the UE can generate and report the first prediction (320) only when one or more channel estimates (i.e., based on aperiodic CSI-RS burst measurements 310) are available within the observation window (330). For example, the first prediction can optionally be a zero-order hold (ZOH) prediction, where the ZOH can be a CSI report for a first observation time instance generated without any prediction. A ZOH prediction is equivalent to 'no prediction', i.e., it is a regular CSI report of the channel measurements. In other words, ZOH can be an observation or measurement of the channel at a time during the observation window (330), and may not be a prediction associated with a point to be predicted that is neither during the observation window (330) nor during the prediction window (355). Optionally, cold start predictions may not be ZOH predictions. Figure 3In the example, the cold start prediction (320) can be determined and reported after two or more measurements or observations (310) of the radio channel at the UE, and therefore may or may not be a ZOH prediction. The cold start prediction can be any prediction made before the first prediction report at n+δ (270), respectively, prior to the CSI report trigger (305). The type and timing of the reported cold start prediction can depend on the UE's NW configuration.

[0039] Alternatively, the UE may also send a prediction for the point to be predicted (315) based on all observations or measurements made during the observation window (330) (e.g., after obtaining the CSI report trigger (305) and during the reporting window (345) between the observation window (330) and the prediction window (350), at intervals d (355). Such predictions may not be considered cold start predictions because they can be made using the complete dataset obtained during the observation window (330).

[0040] The resulting prediction error (340) of the cold start prediction (320) (e.g., squared generalized cosine similarity (SGCS)) can depend on the temporal variance of the radio channel. For the first cold start prediction and / or ZOH prediction, the estimation error can typically be strong or large, as indicated by the magnitude of the vertical arrow of the estimation error (340). These errors may decrease with increasing number of channel observations (310), at least under the assumption of an advanced channel predictor that utilizes all available CSI information from the observation window (330). Note that the estimation error (340) can be a measure of the resulting precoding error or related to intermediate key performance indicators (KPIs) (e.g., SGCS) used for channel prediction quality.

[0041] In example embodiments, the channel predictor (e.g., a UE or AI / ML model) can consider first available channel observations when calculating subsequent channel predictions, which can be expected to result in slightly improved channel predictions because more information is available. An adaptive AI / ML channel predictor can be used to perform channel prediction. Accordingly, the resulting prediction error can be reduced to some extent from left to right, as... Figure 3 As shown, the later cold start prediction (320) is associated with a smaller vertical arrow (340), indicating a reduction in the estimation error.

[0042] In an example embodiment, CSI reporting can begin at the start of the observation window (e.g., cold start prediction). This can have the technical effect of allowing the gNB to pre-encode and transmit user data without any additional latency, even if the resulting prediction accuracy is relatively low. For each further CSI report, all available CSI information to date can be considered, which can have the technical effect of achieving slightly improved CSI predictions (e.g., increased accuracy relative to cold start predictions). For the final CSI report, the length of the observation window can be almost the full observation length pre-configured by the network, such that the resulting prediction performance is almost the best possible performance.

[0043] In the example embodiment, a CSI prediction mode can be configured to include cold start predictions during the observation window.

[0044] The technical effect of the exemplary embodiments of this disclosure is that, compared to conventional operation without channel prediction, it allows the gNB to start user data transmission without any delay.

[0045] The technical effect of the exemplary embodiments of this disclosure can be to avoid any additional latency and ensure maximum user throughput, especially in combinations of IP traffic.

[0046] In the example embodiments, all CSI observations and estimates available for a given CSI reporting time instance can be used for channel state prediction during the observation window. The technical effect of the example embodiments of this disclosure is to obtain the best possible channel prediction accuracy for a given number of channel observations. Compared to conventional systems with no channel predictions during the observation window, the technical effect of the example embodiments of this disclosure is to achieve higher user throughput.

[0047] It can be noted that, in particular, the first cold start prediction may have relatively poor Channel Quality Indicator (CQI) quality, which may impact the overall user data rate compared to the case of permanent channel prediction. Depending on the application and current gNB load conditions or other criteria, cold start CSI reports may be more or less useful to the NW. Accordingly, in the example embodiment, the NW may determine whether to activate or trigger cold start mode prediction based on one or more conditions, factors, or criteria (including but not limited to the foregoing).

[0048] In an example embodiment, the NW can activate or trigger cold start mode prediction. For example, control bits can be added or sent for channel prediction, which can instruct the UE to target the prediction window. Each time instance reports channel state information in either cold start mode or regular mode. Alternatively, cold start mode can be activated or triggered via multiple control bits, dedicated signaling, MAC control elements (CE), downlink control information (DCI), lower-layer signaling, etc. Details of cold start prediction, such as the number of reports during the observation window and the type of CQI report, can be configured via, for example, RRC messages.

[0049] In the example implementation, the number of channel observations and estimates used for the next CSI prediction can be increased for each consecutive cold-start CSI report. In the example implementation, different AI / ML models can be trained for each of the varying numbers of observations. The technical advantage of training multiple different AI / ML models is that it can achieve the best possible channel prediction accuracy for a given number of observations, even at the cost of increased complexity in handling multiple ML models. Alternatively, a single scalable ML model can be used for any number of observations, with the technical advantage of potentially providing slightly reduced performance.

[0050] Now for reference Figure 4 It shows something similar to Figure 3 An example of a cold start prediction mode configuration. Figure 3 and Figure 4 Similar tags will be used, and duplicate descriptions will be omitted. Figure 4 The estimation error (340) on the cold start prediction (320) and the estimation error (410) on the CSI reported for the prediction window (350) are shown.

[0051] In the example embodiment, the time instance used to generate and / or transmit cold start predictions (320) does not necessarily have to be consistent with the channel observation instance (310). For example, there may be a time offset between the time instance used for cold start predictions and the channel observation instance. In the example embodiment, the reporting time instance can be selected independently of the CSI RS used for channel estimation.

[0052] It can be noted that, Figure 4 The overall estimation error is shown, which covers the cold-start prediction plus the cold-start prediction for time instances determined during the observation window (340) and the regular prediction for time instances determined during the prediction window (410). In this case, the estimation error may decrease during the cold-start duration (e.g., 320) due to the increase in the number of channel observations, and may increase slightly with increasing prediction time (e.g., 410) due to the decrease in prediction accuracy. Note that for the prediction window, we assume the full-length observation window and look-ahead prediction. This is a time step. This is similar to traditional channel prediction, where the UE first observes the radio channel within the triggered observation window and uses these observations to infer... Predicted CSI for each prediction time step.

[0053] Varying prediction / estimation errors can result in varying CQIs and varying optimal modulation and coding schemes (MCS) for each CSI report. For example, if the gNB receives a low CQI for cold start prediction, it can use the cold start prediction to transmit DL data with a low MCS or data rate (e.g., pre-coded). The gNB can determine whether and when to use the cold start prediction for DL-based transmissions. In an example embodiment, the UE can report different CQI values ​​for each CSI report. The technical effect is to provide maximum flexibility in reporting the quality of cold start channel predictions at the cost of some overhead.

[0054] In another example embodiment, the UE may report a CQI for the worst estimation error of a first ZOH prediction or a first cold start prediction; another CQI value for the last prediction point (rightmost prediction) made during the observation window; and a CQI value for the minimum prediction error at a first time instance within the prediction window (e.g., the CQI value at the time instance estimated to have the minimum prediction error). Alternatively, a third CQI value may be a CQI value associated with a cold start channel prediction for the last time instance during the prediction window, which may be expected to have a higher error than the prediction for the first time instance during the prediction window (e.g., see the increased error at 410). Based on these three CQI values, the gNB may then interpolate CQI values ​​for other received cold start predictions. For example, the gNB may determine a valid CQI based on three (or more) CQI values ​​received from the UE. For example, the gNB may perform spline interpolation of intermediate CQI values.

[0055] In an example implementation, such a report can be efficiently compressed into multiple incremental CQI values. For instance, after reporting the first CQI, subsequent CQIs can be indicated by the UE using the difference between the first reported CQI and a subsequently determined CQI. In other words, the UE can report only the increments from previously reported CQIs.

[0056] In the example embodiment, the UE can infer the expected prediction performance for cold start prediction. During the observation window, the UE can check one or more previous CSI prediction reports. It can then adapt the CQI report at or near the end of the observation window. In the example embodiment, the UE can report the CQI as a single report at the beginning of the observation window or the cold start prediction report, and can also send an update to the CQI at the end of the observation window.

[0057] In the example embodiment, the UE can perform cold-start channel prediction during the observation window. The technical effect of this example embodiment can be to avoid the associated latency of DL data transmission.

[0058] In the example embodiment, the UE may include all available channel estimates during the cold start period. The technical effect of this example embodiment is that it can achieve the best possible prediction performance / prediction for a given number of channel estimates within the observation window.

[0059] In an example embodiment, control information bits can be defined to allow the NW to switch the UE between having and not having cold start prediction reports. In another example embodiment, the NW can flexibly activate or trigger the cold start prediction mode based on application, cell load, or other criteria using the cold start control bit message. In yet another example embodiment, the NW can use more than one bit to activate different cold start prediction configurations, which can vary in the number of cold start CSI reports, the time instances of cold start CSI reports, etc.

[0060] In example embodiments, the RRC message can be defined to configure cold start prediction details, such as the timing of the first cold start report, the number of cold start reports during the observation window, and the reporting method for CQI values. For example, the UE can be configured to start by sending a ZOH based on a first observation, or it can be configured to start by making a certain number of observations before making a first prediction. For example, the UE can be configured to make a prediction after each consecutive observation or after two or more observations. For example, the UE can be configured to make predictions at a defined period and / or frequency. For example, the UE can be configured to include a CQI index value in each CSI report during the observation window, or to include a CQI index value in a subset of CSI reports during the observation window (e.g., a subset configured to indicate changes in CQI quality over time). For example, the UE can be configured to switch between Type I and Type II reports (e.g., Type I can be used for earlier CSI reports, while Type II can be used for later CSI reports).

[0061] As an alternative to RRC messages, cold start prediction mode configuration can be provided via DCI, MAC CE, low-level signaling, etc. This can be configured as an extension of the general CSI RS configuration message.

[0062] The technical effect of the exemplary embodiments of this disclosure can be to achieve efficient use of resources, that is, to avoid the need to stop data transmission during the observation window.

[0063] The technical effect of the exemplary embodiments of this disclosure can be to achieve low-latency data transmission similar to the case without channel prediction.

[0064] Now for reference Figure 5 A flowchart relating to gNB actions related to cold start prediction operations according to an example embodiment of this disclosure is shown. At 510, the gNB may configure the UE for cold start prediction. For example, the gNB may send one or more configuration details to the UE for performing cold start prediction. At 520, the gNB may send, for example, activation bits or configurations for cold start prediction via the Physical Downlink Control Channel (PDCCH), MAC CE, DCI messages, lower-layer signaling, etc. The gNB may activate configurations previously provided to the UE. At 530, the gNB may send a CSI RS. At 540, the gNB may receive a cold start CSI, such as a CSI report received during the UE observation window. The cold start CSI may be channel quality information / prediction at a point in time during the observation window, or channel quality prediction at a point in time during the prediction window. At 550, the gNB may precode DL data based at least in part on the cold start CSI. The DL data may be sent with a delay after receiving the CSI report. For example, the DL data may be sent during or after the observation window. At 560, gNB can receive predicted CSI (e.g., CSI after the observation window is completed) and precode DL data using predicted Type II Doppler CSI.

[0065] Now for reference Figure 6 The diagram illustrates a flowchart relating to UE actions related to cold start prediction operations according to an example embodiment of this disclosure. At 610, the UE may activate cold start prediction. For example, the UE may activate a previously received configuration based on an indication received from the NW. At 620, the UE may, for example, receive the CSI RS and estimate the radio channel during the observation window. At 620, the UE may report the cold start CSI and optionally the CQI. At 630, the UE may infer the predicted CSI (e.g., based on all measurements made during the observation window). The CSI can be predicted for a point in time or instance during the observation window and / or the prediction window. At 640, the UE may report the predicted CSI and CQI.

[0066] Now for reference Figure 7This diagram illustrates a flowchart related to UE and NW cold start prediction operations according to an example embodiment of this disclosure. At 705, the gNB can define a cold start prediction mode offline or via RRC. At 710, the gNB can send activation of the cold start prediction mode to the UE. At 715, the gNB can send the CSI RS for each AP to the UE. At 720, the UE can estimate the CSI during the observation window. The estimated or predicted CSI can be used for time instances during the observation window and / or time instances during the prediction window. The CSI predicted for time instances in the prediction window can be based on all observations made during the observation window. At 725, the UE can report the CSI cold start prediction to the gNB. For example, ZOH can be reported. The report can be configured according to the cold start prediction. At 730, the UE can report the cold start CQI to the gNB. The CQI can be reported together with the cold start prediction or it can be reported separately. The CQI can be reported at a different frequency than the cold start prediction. Steps 720, 725, and 730 can be repeated multiple times during the observation window. At 735, the gNB can use the cold-start prediction for multiple-input multiple-output (MIMO) precoding and / or MCS adaptation, for example, with a delay after receiving the CSI report. Step 725 can be performed during or after the observation window. At 740, the UE can predict the CSI for the prediction window, for example, after the observation window ends. At 745, the UE can report the predicted CSI for the prediction window. At 750, the gNB can use the predicted CSI for MIMO precoding and / or MCS adaptation.

[0067] Figure 8 Potential steps of example method 800 are illustrated. Example method 800 may include: determining to perform channel prediction and reporting during an observation window, 810; determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being at least partially based on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window, 820; and transmitting at least one of the at least one channel prediction or at least one channel observation determined during the observation window, 830. Example method 800 may be performed, for example, by a UE or a smartphone device.

[0068] Figure 9Potential steps of example method 900 are illustrated. Example method 900 may include: sending an instruction to a user equipment to perform channel prediction and reporting during an observation window, 910; receiving at least one of the following during the observation window: at least one channel prediction made during the observation window, or at least one channel observation made during the observation window, 920; and transmitting downlink data at least in part based on at least one of the at least one channel prediction or at least one channel observation, 930. Example method 900 may be performed, for example, by a network node, base station, gNB, network entity, network function, etc.

[0069] According to one example embodiment, an apparatus may include: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: determine to perform channel prediction and reporting during an observation window; determine at least one of the following during the observation window: at least one channel prediction, which is at least partially based on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window; and transmit at least one of the at least one channel prediction or at least one of the at least one channel observations during the observation window. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances during the observation window. The at least one observation of the channel may include a zero-order hold prediction. Determining to perform channel prediction and reporting during the observation window may include: the example apparatus is further configured to: receive an instruction to perform channel prediction and reporting during the observation window. The instruction may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. The example apparatus can also be configured to receive a configuration for performing channel prediction and reporting during an observation window, wherein the configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction, the number of channel predictions to be made and reported during the observation window, a period for at least one channel prediction, a frequency for at least one channel prediction, or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. This configuration may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window, or at least one channel prediction for at least one time instance during the observation window. The example apparatus may also be configured to transmit at least one channel quality indicator during an observation window, wherein one of the following is true: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one determined channel prediction; or the at least one channel quality indicator may include at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for the last time instance during a prediction window period following the observation window.At least one of the determined channel predictions or at least one channel measurement may include multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The at least one channel prediction may be determined using at least one learning-based model. The at least one learning-based model may include multiple learning-based models, wherein a corresponding learning-based model among the multiple learning-based models may be configured to take observations of a different number of channels as input. The at least one learning-based model may include a scalable learning-based model.

[0070] According to an example embodiment, an example method may be provided, comprising: determining, using a user equipment, to perform channel prediction and reporting during an observation window; determining, during the observation window, at least one of the following: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window; and transmitting, during the observation window, at least one of the at least one channel prediction or at least one channel observation determined therein. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances within the observation window. The at least one observation of the channel may include a zero-order hold prediction. Determining to perform channel prediction and reporting during the observation window may include: receiving an instruction to perform channel prediction and reporting during the observation window. The instruction may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. The exemplary method may further include: receiving a configuration for performing channel prediction and reporting during an observation window, wherein the configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction, the number of channel predictions to be made and reported during the observation window, a period for at least one channel prediction, a frequency for at least one channel prediction, or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. The configuration may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window, or at least one channel prediction for at least one time instance during the observation window. The example method may further include: transmitting at least one channel quality indicator during an observation window, wherein one of the following is true: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one determined channel prediction; or the at least one channel quality indicator may include at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for a last time instance during a prediction window following the observation window. At least one of the determined at least one channel prediction or at least one channel measurement may include multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The at least one channel prediction may be determined using at least one learning-based model.At least one learning-based model may include multiple learning-based models, wherein a corresponding learning-based model among the multiple learning-based models may be configured to take observations from a different number of channels as input. At least one learning-based model may include scalable learning-based models.

[0071] According to one example embodiment, an apparatus may include: circuitry configured to perform: determining, using a user equipment, channel prediction and reporting during an observation window; circuitry configured to perform: determining, during the observation window, at least one of the following: at least one channel prediction, the at least one channel prediction being at least partially based on one or more observations of a channel made during the observation window, or at least one observation of a channel made during the observation window; and circuitry configured to perform: transmitting, during the observation window, at least one of the determined at least one channel prediction or at least one channel observation.

[0072] According to one example embodiment, an apparatus may include: processing circuitry; and memory circuitry including computer-readable code, the memory circuitry and the computer-readable code being configured to enable the apparatus, via the processing circuitry, to: determine to perform channel prediction and reporting during an observation window; determine at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of a channel made during the observation window; and transmit at least one of the determined at least one channel prediction or the at least one channel observation during the observation window.

[0073] As used herein, the terms “circuit system” or “component” may refer to one or more or all of the following: (a) a hardware circuit implementation (such as an implementation in an analog, digital, and / or quantum circuit system); and (b) a combination of (multiple) hardware circuits and software, such as (if applicable): (i) a combination of (multiple) analog, digital, and / or quantum hardware circuits with software / firmware; and (ii) any or all portions of (multiple) hardware processors having software (including (multiple) digital and / or quantum processors), and (multiple) memories that work together to enable a device (such as a mobile device, computing device, or server) to perform various functions; and (c) any or all portions of (multiple) hardware circuits that require software (e.g., firmware) for operation, such as (multiple) microprocessors, (multiple) processors, and / or (multiple) quantum processors, but where the software may be absent when it is not required for operation. This definition of circuit applies to all use of the term herein (including in any claim). As a further example, as used herein, the term "circuit" also encompasses only hardware circuitry or a processor (or multiple processors), or portions of hardware circuitry or a processor and their accompanying software and / or firmware. For example, where applicable to certain claim elements, the term "circuit" also encompasses baseband integrated circuits or processor integrated circuits for mobile devices, or similar integrated circuits in servers, cellular network devices, or other computing or network devices.

[0074] According to one example embodiment, an apparatus may include components for causing the apparatus to at least perform: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being at least partially based on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window; and transmitting said at least one of the determined at least one channel prediction or at least one channel observation during the observation window. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances during the observation window. The at least one observation of the channel may include a zero-order hold prediction. Components configured to cause the apparatus to perform determining to perform channel prediction and reporting during the observation window may include components for causing the apparatus to perform: receiving an instruction to perform channel prediction and reporting during the observation window. This instruction may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. The component can also be configured to cause the apparatus to perform: receiving a configuration for performing channel prediction and reporting during an observation window, wherein the configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction, the number of channel predictions to be made and reported during the observation window, a period for at least one channel prediction, a frequency for at least one channel prediction, or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. This configuration may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window, or at least one channel prediction for at least one time instance during the observation window. The component can also be configured to cause the apparatus to perform: transmitting at least one channel quality indicator during an observation window, wherein one of the following is true: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one determined channel prediction; or the at least one channel quality indicator may include at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for the last time instance during a prediction window period following the observation window.At least one of the determined channel predictions or at least one channel measurement may include multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The at least one channel prediction may be determined using at least one learning-based model. The at least one learning-based model may include multiple learning-based models, wherein a corresponding learning-based model among the multiple learning-based models may be configured to take observations of a different number of channels as input. The at least one learning-based model may include a scalable learning-based model.

[0075] Processors, memory, and / or example algorithms (which may be encoded as instructions, programs, or code) may be provided as example components for providing or causing the execution of operations.

[0076] According to one example embodiment, a (non-transitory) computer-readable medium includes instructions stored thereon that, when executed by at least one processor, cause at least one processor to: determine to perform channel prediction and reporting during an observation window; determine at least one of the following during the observation window: at least one channel prediction, which is at least partially based on one or more observations of a channel made during the observation window, or at least one observation of a channel made during the observation window; and cause at least one of the determined at least one channel prediction or at least one channel observation to be transmitted during the observation window.

[0077] According to one example embodiment, a (non-transitory) computer-readable medium includes instructions stored thereon for performing at least the following: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, which is at least partially based on one or more observations of the channel made during the observation window, or at least one observation of the channel made during the observation window; and causing to transmit at least one of the at least one channel prediction or at least one channel observation determined during the observation window. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances during the observation window. The at least one observation of the channel may include a zero-order hold prediction. The instructions stored thereon for performing the determination to perform channel prediction and reporting during the observation window may include instructions for performing: receiving an instruction to perform channel prediction and reporting during the observation window. This instruction may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. The example computer-readable medium may also include instructions stored thereon for performing: causing the receipt of a configuration for performing channel prediction and reporting during an observation window, wherein the configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction, the number of channel predictions to be made and reported during the observation window, a period for at least one channel prediction, a frequency for at least one channel prediction, or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. The configuration may be received via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window, or at least one channel prediction for at least one time instance during the observation window. The example computer-readable medium may also include instructions stored thereon for performing: causing at least one channel quality indicator to be transmitted during an observation window, wherein one of the following: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one determined channel prediction; or the at least one channel quality indicator may include at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for the last time instance during a prediction window period following the observation window.At least one of the determined channel predictions or at least one channel measurement may include multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The at least one channel prediction may be determined using at least one learning-based model. The at least one learning-based model may include multiple learning-based models, wherein a corresponding learning-based model among the multiple learning-based models may be configured to take observations of a different number of channels as input. The at least one learning-based model may include a scalable learning-based model.

[0078] According to one example embodiment, a machine-readable (non-transitory) program storage device may be provided, tangibly embodying machine-executable instructions for performing operations including: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of a channel made during the observation window; and causing at least one of the at least one channel prediction or at least one channel observation determined during the observation window to be transmitted.

[0079] According to one example embodiment, a (non-transitory) computer-readable medium includes instructions that, when executed by a device, cause the device to perform at least the following: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of the channel made during the observation window; and causing at least one of the at least one channel prediction or at least one channel observation determined during the observation window to be transmitted.

[0080] According to one example embodiment, a computer-implemented system includes: at least one processor and at least one (non-transitory) memory storing instructions, the instructions, when executed by the at least one processor, causing the system to at least: determine to perform channel prediction and reporting during an observation window; determine at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of said channel made during the observation window; and cause to transmit at least one of the at least one channel prediction or said at least one channel observation determined during the observation window.

[0081] According to one example embodiment, a computer-implemented system includes: components for determining channel prediction and reporting to be performed during an observation window; components for determining at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of a channel made during the observation window; and components for causing at least one of the at least one determined in the at least one channel prediction or at least one channel observation to be transmitted during the observation window.

[0082] According to one example embodiment, an apparatus may include: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: send an instruction to a user equipment to perform channel prediction and reporting during an observation window; receive, during the observation window, at least one of: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and transmit downlink data at least in part based on at least one of the at least one channel prediction or at least one channel observation. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances within the observation window. The at least one observation of the channel may include a zero-order hold prediction. The instruction to perform channel prediction and reporting may be sent via at least one of: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling.

[0083] The example apparatus can also be configured to send a configuration to a user equipment for performing channel prediction and reporting during an observation window, wherein the indication for performing channel prediction and reporting during the observation window may include activation of the configuration. The configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction; the number of channel predictions to be made and reported during the observation window; a period for at least one channel prediction; a frequency for at least one channel prediction; or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. The configuration may be sent via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window, or at least one channel prediction for at least one time instance during the observation window. The example apparatus can also be configured to receive at least one channel quality indicator during an observation window, wherein one of the following is true: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one channel prediction; or the at least one channel quality indicator may include at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for a last time instance during a prediction window following the observation window, wherein downlink data may also be transmitted based on at least one channel quality indicator. The at least one channel quality indicator may include a first channel quality indicator, a second channel quality indicator, and a third channel quality indicator, wherein the example apparatus may also be configured to determine a valid channel quality indicator at least in part based on the first channel quality indicator, the second channel quality indicator, and the third channel quality indicator. At least one of the channel predictions or at least one channel measurement may include multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The example apparatus may also be configured to determine whether to activate channel prediction and reporting during the observation window, at least in part based on at least one of downlink data or apparatus load conditions.

[0084] According to an example embodiment, an example method may be provided, comprising: sending an instruction to a user equipment via a network node to perform channel prediction and reporting during an observation window; receiving, during the observation window, at least one of the following: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and transmitting downlink data at least in part based on at least one of the at least one channel prediction or at least one channel observation. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances within the observation window. The at least one observation of the channel may include a zero-order hold prediction. The instruction to perform channel prediction and reporting may be sent via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. The example method may further include: sending a configuration to the user equipment for performing channel prediction and reporting during the observation window, wherein the instruction to perform channel prediction and reporting during the observation window may include activation of the configuration. The configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction; the number of channel predictions to be made and reported during the observation window; a period for at least one channel prediction; a frequency for at least one channel prediction; or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. The configuration may be transmitted via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window; or at least one channel prediction for at least one time instance during the observation window. The example method may further include: receiving at least one channel quality indicator during an observation window, wherein one of the following is true: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one channel prediction; or the at least one channel quality indicator includes at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for a last time instance during a prediction window following the observation window, wherein downlink data may also be transmitted based on at least one channel quality indicator. The at least one channel quality indicator may include a first channel quality indicator, a second channel quality indicator, and a third channel quality indicator, wherein the example method may further include: determining a valid channel quality indicator based at least in part on the first channel quality indicator, the second channel quality indicator, and the third channel quality indicator.At least one of the channel predictions or at least one channel measurement may include multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The example method may also include determining whether to activate channel prediction and reporting during the observation window based at least in part on at least one of downlink data or the load status of network nodes.

[0085] According to one example embodiment, an apparatus may include: circuitry configured to perform: sending an instruction to a user equipment via a network node to perform channel prediction and reporting during an observation window; circuitry configured to perform: receiving at least one of the following during the observation window: at least one channel prediction made during the observation window, or at least one channel observation made during the observation period; and circuitry configured to perform: transmitting downlink data based at least in part on at least one of the at least one channel prediction or at least one channel observation.

[0086] According to one example embodiment, an apparatus may include: processing circuitry; and memory circuitry including computer-readable code, the memory circuitry and the computer-readable code being configured to enable the apparatus to: send an instruction to a user equipment to perform channel prediction and reporting during an observation window; receive, during the observation window, at least one of the following: at least one channel prediction made during the observation window, or at least one channel observation made during the observation window; and transmit downlink data based at least in part on at least one of the at least one channel prediction or at least one channel observation.

[0087] According to one example embodiment, an apparatus may include components for causing the apparatus to perform at least the following: sending an instruction to a user equipment to perform channel prediction and reporting during an observation window; receiving at least one of the following during the observation window: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and transmitting downlink data at least in part based on at least one of the at least one channel prediction or at least one channel observation. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances within the observation window. The at least one observation of the channel may include a zero-order hold prediction. The instruction to perform channel prediction and reporting may be sent via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. The components may also be configured to cause the apparatus to perform: sending a configuration to the user equipment for performing channel prediction and reporting during the observation window, wherein the instruction to perform channel prediction and reporting during the observation window may include activation of the configuration. The configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction; the number of channel predictions to be made and reported during the observation window; a period for at least one channel prediction; a frequency for at least one channel prediction; or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. The configuration may be transmitted via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window; or at least one channel prediction for at least one time instance during the observation window. The component can also be configured to cause the apparatus to perform: receiving at least one channel quality indicator during an observation window, wherein one of the following is true: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one channel prediction; or the at least one channel quality indicator may include at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for the last time instance during a prediction window period following the observation window, wherein downlink data may also be transmitted based on at least one channel quality indicator. The at least one channel quality indicator may include a first channel quality indicator, a second channel quality indicator, and a third channel quality indicator, wherein the component can also be configured to cause the apparatus to perform: determining a valid channel quality indicator based at least in part on the first channel quality indicator, the second channel quality indicator, and the third channel quality indicator.At least one of the channel predictions or at least one channel measurement may include multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The component may also be configured to cause the device to perform: determining whether to activate channel prediction and reporting during the observation window, at least in part based on at least one of downlink data or the device's load conditions.

[0088] According to one example embodiment, a (non-transitory) computer-readable medium includes instructions stored thereon that, when executed by at least one processor, cause at least one processor to: send an instruction to a user equipment via a network node to perform channel prediction and reporting during an observation window; cause the processor to receive at least one of the following during the observation window: at least one channel prediction made during the observation window, or at least one observation of a channel made during the observation period; and cause the processor to transmit downlink data based at least in part on at least one of the at least one channel prediction or at least one channel observation.

[0089] According to one example embodiment, a (non-transitory) computer-readable medium includes instructions stored thereon for performing at least the following: causing a network node to send an instruction to a user equipment to perform channel prediction and reporting during an observation window; causing at least one of the following to be received during the observation window: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and causing downlink data to be transmitted at least in part based on at least one of the at least one channel prediction or at least one channel observation. The at least one channel prediction may include a prediction of the channel based on observations of the channel during at least two time instances within the observation window. The at least one observation of the channel may include a zero-order hold prediction. The instruction to perform channel prediction and reporting may be sent via at least one of the following: radio resource control signaling, downlink control information, media access control signaling, or lower-layer signaling. The example computer-readable medium may also include instructions stored thereon for performing: causing a configuration to be sent to the user equipment for performing channel prediction and reporting during the observation window, wherein the instruction to perform channel prediction and reporting during the observation window may include activation of the configuration. The configuration may include at least one of the following: a time instance for reporting a first channel prediction in at least one channel prediction; the number of channel predictions to be made and reported during the observation window; a period for at least one channel prediction; a frequency for at least one channel prediction; or a configuration for reporting at least one channel quality indicator associated with at least one channel prediction. The configuration may be transmitted via at least one of the following: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling. At least one channel prediction may include at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window; or at least one channel prediction for at least one time instance during the observation window. The example computer-readable medium may also include instructions stored thereon for performing: causing at least one channel quality indicator to be received during an observation window, wherein one of the following: each channel quality indicator in the at least one channel quality indicator may be associated with a corresponding channel prediction in at least one channel prediction; or the at least one channel quality indicator may include at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for a last time instance during a prediction window following the observation window, wherein downlink data may also be transmitted based on at least one channel quality indicator.The at least one channel quality indicator may include a first channel quality indicator, a second channel quality indicator, and a third channel quality indicator. The example computer-readable medium may also include instructions stored thereon for performing: determining a valid channel quality indicator based at least in part on the first, second, and third channel quality indicators. At least one of the at least one channel prediction or at least one channel measurement includes multiple channel state information reports, wherein the estimation error associated with a first channel state information report among the multiple channel state information reports may be greater than at least one estimation error associated with at least one second channel state information report among the multiple channel state information reports. The example computer-readable medium may also include instructions stored thereon for performing: determining whether to activate channel prediction and reporting during the observation window based at least in part on at least one of downlink data or the load condition of a network node.

[0090] According to one example embodiment, a machine-readable (non-transitory) program storage device may be provided, tangibly embodying machine-executable instructions for performing operations including: causing a network node to send an instruction to a user equipment to perform channel prediction and reporting during an observation window; causing at least one of the following to be received during the observation window: at least one channel prediction made during the observation window, or at least one observation of a channel made during the observation window; and causing downlink data to be transmitted at least in part based on at least one of the at least one channel prediction or at least one channel observation.

[0091] According to one example embodiment, a (non-transitory) computer-readable medium includes instructions that, when executed by a device, cause the device to perform at least the following: cause a network node to send an instruction to a user equipment to perform channel prediction and reporting during an observation window; cause the device to receive at least one of the following during the observation window: at least one channel prediction made during the observation window, or at least one channel observation made during the observation window; and cause downlink data to be transmitted based at least in part on at least one of the at least one channel prediction or at least one channel observation.

[0092] According to one example embodiment, a computer-implemented system includes: at least one processor and at least one (non-transitory) memory storing instructions, the instructions, when executed by the at least one processor, causing the system to at least: send an instruction to a user equipment via a network node to perform channel prediction and reporting during an observation window; receive at least one of the following during the observation window: at least one channel prediction made during the observation window, or at least one observation of a channel made during the observation window; and transmit downlink data at least in part based on at least one of the at least one channel prediction or at least one channel observation.

[0093] According to one example embodiment, a computer-implemented system includes: components for causing a network node to send an instruction to a user equipment to perform channel prediction and reporting during an observation window; components for causing at least one of the following to be received during the observation window: at least one channel prediction made during the observation window, or at least one channel observation made during the observation window; and components for causing downlink data to be transmitted based at least in part on at least one of the at least one channel prediction or at least one channel observation.

[0094] Furthermore, the various implementations of this disclosure can be described with reference to the following terms, and their features can be combined in any reasonable manner.

[0095] Clause 1. An apparatus for communication, comprising: at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the apparatus to at least: determine to perform channel prediction and reporting during an observation window; determine at least one of the following during the observation window: at least one channel prediction, the at least one channel prediction being at least partially based on one or more observations of a channel made during the observation window, or at least one observation of the channel made during the observation window; and transmit the at least one determined in the at least one channel prediction or the at least one observation of the channel during the observation window.

[0096] Clause 2. The apparatus according to Clause 1, wherein the at least one channel prediction comprises a prediction of the channel based on observations of the channel during at least two time instances during the observation window.

[0097] Clause 3. The apparatus according to Clause 1 or 2, wherein the at least one observation of the channel includes zero-order hold prediction.

[0098] Clause 4. The apparatus according to any one of Clauses 1 to 3, wherein determining to perform channel prediction and reporting during the observation window includes, when the instruction is executed by the at least one processor, causing the apparatus to: receive an instruction to perform channel prediction and reporting during the observation window.

[0099] Clause 5. The apparatus according to Clause 4, wherein the instruction is received via at least one of: radio resource control signaling, downlink control information, media access control signaling, or lower-layer signaling.

[0100] Clause 6. An apparatus according to any one of Clauses 1 to 5, wherein the instructions, when executed by the at least one processor, cause the apparatus to: receive a configuration for performing channel prediction and reporting during the observation window, wherein the configuration includes at least one of: a time instance for reporting a first channel prediction in the at least one channel prediction, the number of channel predictions to be made and reported during the observation window, a period for the at least one channel prediction, a frequency for the at least one channel prediction, or a configuration for reporting at least one channel quality indicator associated with the at least one channel prediction.

[0101] Clause 7. The apparatus according to Clause 6, wherein the configuration is received via at least one of: radio resource control signaling, downlink control information, media access control signaling, or lower-layer signaling.

[0102] Clause 8. The apparatus according to any one of Clauses 1 to 7, wherein the at least one channel prediction comprises at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window, or at least one channel prediction for at least one time instance during the observation window.

[0103] Clause 9. An apparatus according to any one of Clauses 1 to 8, wherein the instructions, when executed by the at least one processor, cause the apparatus to: transmit at least one channel quality indicator during the observation window, wherein one of the at least one channel quality indicator is associated with a corresponding channel prediction in the at least one determined channel prediction; or the at least one channel quality indicator includes at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for a last time instance during a prediction window period following the observation window.

[0104] Clause 10. The apparatus according to any one of Clauses 1 to 9, wherein the at least one determined in the at least one channel prediction or the at least one observation of the channel comprises a plurality of channel state information reports, wherein the estimation error associated with a first channel state information report among the plurality of channel state information reports is greater than at least one estimation error associated with at least one second channel state information report among the plurality of channel state information reports.

[0105] Clause 11. The apparatus according to any one of Clauses 1 to 10, wherein the at least one channel prediction is determined using at least one learning-based model.

[0106] Clause 12. The apparatus according to Clause 11, wherein the at least one learning-based model comprises a plurality of learning-based models, wherein a corresponding learning-based model among the plurality of learning-based models is configured to take a different number of observations of the channel as input.

[0107] Clause 13. The apparatus according to Clause 11, wherein the at least one learning-based model includes a scalable learning-based model.

[0108] Clause 14. A method for communication, comprising: using a user equipment to determine performing channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, said at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of the channel made during the observation window; and transmitting said at least one of the at least one channel prediction or the at least one observation of the channel determined during the observation window.

[0109] Clause 15. An apparatus for communication, comprising components for causing the apparatus to perform at least: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, said at least one channel prediction being at least partially based on one or more observations of a channel made during the observation window, or at least one observation of the channel made during the observation window; and transmitting said at least one of the at least one channel prediction or the at least one observation of the channel made during the observation window.

[0110] Clause 16. A computer-readable medium comprising instructions stored thereon for performing at least the following: determining to perform channel prediction and reporting during an observation window; determining at least one of the following during the observation window: at least one channel prediction, said at least one channel prediction being based at least in part on one or more observations of a channel made during the observation window, or at least one observation of the channel made during the observation window; and causing said at least one of the at least one channel prediction or the at least one observation of the channel to be transmitted during the observation window.

[0111] Clause 17. An apparatus for communication, comprising: at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the apparatus to at least: send to a user equipment an instruction to perform channel prediction and reporting during an observation window; receive during the observation window at least one of: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and transmit downlink data at least in part based on at least one of the at least one channel prediction or the at least one observation of the channel.

[0112] Clause 18. The apparatus according to Clause 17, wherein the at least one channel prediction comprises a prediction of the channel based on observations of the channel during at least two time instances during the observation window.

[0113] Clause 19. The apparatus according to Clause 17 or 18, wherein the at least one observation of the channel includes zero-order hold prediction.

[0114] Clause 20. The apparatus according to any one of Clauses 17 to 19, wherein the instruction to perform channel prediction and reporting is transmitted via at least one of: radio resource control signaling, downlink control information, medium access control signaling, or lower-layer signaling.

[0115] Clause 21. The apparatus according to any one of Clauses 17 to 20, wherein the instructions, when executed by the at least one processor, cause the apparatus to: send to the user equipment a configuration for performing channel prediction and reporting during the observation window, wherein the instruction to perform channel prediction and reporting during the observation window includes activation of the configuration.

[0116] Clause 22. The apparatus according to Clause 21, wherein the configuration includes at least one of the following: a time instance for reporting a first channel prediction in the at least one channel prediction, the number of channel predictions to be made and reported during the observation window, a period for the at least one channel prediction, a frequency for the at least one channel prediction, or a configuration for reporting at least one channel quality indicator associated with the at least one channel prediction.

[0117] Clause 23. The apparatus according to Clause 21 or 22, wherein said configuration is transmitted via at least one of: radio resource control signaling, downlink control information, media access control signaling, or lower-layer signaling.

[0118] Clause 24. The apparatus according to any one of Clauses 17 to 23, wherein the at least one channel prediction comprises at least one of the following: at least one channel prediction for at least one time instance during a prediction window following the observation window, or at least one channel prediction for at least one time instance during the observation window.

[0119] Clause 25. An apparatus according to any one of Clauses 17 to 24, wherein the instructions, when executed by the at least one processor, cause the apparatus to: receive at least one channel quality indicator during the observation window, wherein one of the at least one channel quality indicator is associated with a corresponding channel prediction in the at least one channel prediction; or the at least one channel quality indicator includes at least: a first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for a last time instance during a prediction window following the observation window, wherein the downlink data is also transmitted based on the at least one channel quality indicator.

[0120] Clause 26. The apparatus of Clause 25, wherein the at least one channel quality indicator comprises a first channel quality indicator, a second channel quality indicator, and a third channel quality indicator, wherein the instructions, when executed by the at least one processor, cause the apparatus to: determine a valid channel quality indicator based at least in part on the first channel quality indicator, the second channel quality indicator, and the third channel quality indicator.

[0121] Clause 27. The apparatus according to any one of Clauses 17 to 26, wherein the at least one of the at least one channel prediction or at least one observation of the channel comprises a plurality of channel state information reports, wherein an estimation error associated with a first channel state information report among the plurality of channel state information reports is greater than at least one estimation error associated with at least one second channel state information report among the plurality of channel state information reports.

[0122] Clause 28. An apparatus according to any one of Clauses 17 to 27, wherein the instructions, when executed by the at least one processor, cause the apparatus to determine whether to activate channel prediction and reporting during the observation window based at least in part on at least one of the following: the downlink data, or the load status of the apparatus.

[0123] Clause 29. A method for communication, comprising: transmitting to a user equipment via a network node an instruction to perform channel prediction and reporting during an observation window; receiving during the observation window at least one of: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and transmitting downlink data at least in part based on at least one of the at least one channel prediction or the at least one observation of the channel.

[0124] Clause 30. An apparatus for communication, comprising components for causing the apparatus to perform at least: sending an instruction to a user equipment to perform channel prediction and reporting during an observation window; receiving during the observation window at least one of: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and transmitting downlink data at least in part based on at least one of the at least one channel prediction or the at least one observation of the channel.

[0125] Clause 31. A computer-readable medium comprising instructions stored thereon, the instructions being configured to perform at least the following: causing a network node to send an instruction to a user equipment to perform channel prediction and reporting during an observation window; causing at least one of the following to be received during the observation window: at least one channel prediction made during the observation window, or at least one observation of the channel made during the observation window; and causing downlink data to be transmitted at least in part based on at least one of the at least one channel prediction or the at least one observation of the channel.

[0126] As used herein, the term “non-transient” refers to a limitation on the medium itself (i.e., tangible, not signaling), rather than a limitation on the persistence of data storage (e.g., RAM vs. ROM).

[0127] As used herein, the terms “at least one” and “one or more” mean “any one of at least one” and “any one of one or more”, respectively.

[0128] It should be understood that the foregoing description is illustrative only. Those skilled in the art can devise various alternatives and modifications. For example, the features recited in the various dependent claims can be combined with each other in any suitable combination. Furthermore, features from the different embodiments described above can be selectively combined to form new embodiments. Accordingly, this specification is intended to cover all such alternatives, modifications, and variations falling within the scope of the appended claims.

Claims

1. A device for communication, comprising: At least one processor; as well as At least one memory, the at least one memory storing instructions, the instructions, when executed by the at least one processor, cause the device to at least: Determine whether to perform channel prediction and reporting during the observation window; During the observation window, at least one of the following must be determined: At least one channel prediction, said at least one channel prediction being based at least in part on one or more channel observations made during said observation window, or At least one observation of the channel made during the observation window; and During the observation window, transmit the at least one channel prediction or the at least one determined in the at least one observation of the channel.

2. The apparatus of claim 1, wherein the at least one channel prediction comprises a prediction of the channel based on observations of the channel during at least two time instances within the observation window.

3. The apparatus of claim 1, wherein the at least one observation of the channel comprises zero-order hold prediction.

4. The apparatus of claim 1, wherein determining to perform channel prediction and reporting during the observation window includes, when the instruction is executed by the at least one processor, causing the apparatus to: Receive instructions to perform channel prediction and reporting during the observation window.

5. The apparatus of claim 4, wherein the instruction is received via at least one of the following: Radio resource control signaling, Downlink control information, Media access control signaling, or Low-level signaling.

6. The apparatus of claim 1, wherein the instructions, when executed by the at least one processor, cause the apparatus to: Receive a configuration for performing channel prediction and reporting during the observation window, wherein the configuration includes at least one of the following: Used to report the time instance of the first channel prediction in the at least one channel prediction. The number of channel predictions to be made and reported during the observation window. For the period of the at least one channel prediction, The frequency used for prediction of the at least one channel, or Configuration for reporting at least one channel quality indicator associated with the at least one channel prediction.

7. The apparatus of claim 6, wherein the configuration is received via at least one of the following: Radio resource control signaling, Downlink control information, Media access control signaling, or Low-level signaling.

8. The apparatus of claim 1, wherein the at least one channel prediction comprises at least one of the following: At least one channel prediction for at least one time instance during the prediction window following the observation window, or At least one channel prediction for at least one time instance during the observation window.

9. The apparatus of claim 1, wherein the instructions, when executed by the at least one processor, cause the apparatus to: At least one channel quality indicator is transmitted during the observation window, wherein one of the following is true: Each of the at least one channel quality indicator is associated with a corresponding channel prediction in the at least one determined channel prediction; or The at least one channel quality indicator includes at least: A first channel quality indicator associated with a first channel prediction or observation made during the observation window, a second channel quality indicator associated with a last channel prediction made during the observation window, and a third channel quality indicator associated with a channel prediction for the last time instance during a prediction window following the observation window.

10. The apparatus of any one of claims 1 to 9, wherein the at least one determined in the at least one channel prediction or the at least one observation of the channel comprises a plurality of channel state information reports, wherein an estimation error associated with a first channel state information report among the plurality of channel state information reports is greater than at least one estimation error associated with at least one second channel state information report among the plurality of channel state information reports.

Images