Systems and methods of uplink control information transmission and / or retransmission for artificial intelligence / machine learning based time- spatial-frequency domain compression
Time-spatial-frequency domain AI/ML models with synchronized encoders and decoders address inefficiencies in CSI compression by incorporating time domain aspects, enhancing performance and reducing errors in wireless communication systems.
Patent Information
- Application Number
- PCT/CN2024/077284
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-02-16
- Publication Date
- 2025-08-21
AI Technical Summary
Existing wireless communication systems do not effectively utilize time domain aspects in artificial intelligence/machine learning (AI/ML) models for channel state information (CSI) compression, leading to inefficiencies and potential forward error propagation issues.
Implementing time-spatial-frequency domain AI/ML models that include encoders and decoders with internal states to account for time domain aspects, along with procedures for triggering, resetting, and retransmission of CSI feedback to maintain synchronization between the UE and base station.
Enhances CSI feedback performance by reducing complexity and overhead, improving alignment between encoder and decoder states, and preventing forward error propagation.
Smart Images

Figure CN2024077284_21082025_PF_FP_ABST
Abstract
Description
[Rectified under Rule 91, 24.09.2024]SYSTEMS AND METHODS OF UPLINK CONTROL INFORMATION TRANSMISSION AND / OR RETRANSMISSION FOR ARTIFICIAL INTELLIGENCE / MACHINE LEARNING BASED TIME-SPATIAL-FREQUENCY DOMAIN COMPRESSIONTECHNICAL FIELD
[0001] This application relates generally to wireless communication systems, including wireless communication systems that implement artificial intelligence (AI) / machine learning (ML) models.BACKGROUND
[0002] Wireless mobile communication technology uses various standards and protocols to transmit data between a base station and a wireless communication device. Wireless communication system standards and protocols can include, for example, 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) (e.g., 4G) , 3GPP New Radio (NR) (e.g., 5G) , and Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard for Wireless Local Area Networks (WLAN) (commonly known to industry groups as ) .
[0003] As contemplated by the 3GPP, different wireless communication systems's tandards and protocols can use various radio access networks (RANs) for communicating between a base station of the RAN (which may also sometimes be referred to generally as a RAN node, a network node, or simply a node) and a wireless communication device known as a user equipment (UE) . 3GPP RANs can include, for example, Global System for Mobile communications (GSM) , Enhanced Data Rates for GSM Evolution (EDGE) RAN (GERAN) , Universal Terrestrial Radio Access Network (UTRAN) , Evolved Universal Terrestrial Radio Access Network (E-UTRAN) , and / or Next-Generation Radio Access Network (NG-RAN) .
[0004] Each RAN may use one or more radio access technologies (RATs) to perform communication between the base station and the UE. For example, the GERAN implements GSM and / or EDGE RAT, the UTRAN implements Universal Mobile Telecommunication System (UMTS) RAT or other 3GPP RAT, the E-UTRAN implements LTE RAT (sometimes simply referred to as LTE) , and NG-RAN implements NR RAT (sometimes referred to herein as 5G RAT, 5G NR RAT, or simply NR) . In certain deployments, the E-UTRAN may also implement NR RAT. In certain deployments, NG-RAN may also implement LTE RAT.
[0005] A base station used by a RAN may correspond to that RAN. One example of an E-UTRAN base station is an Evolved Universal Terrestrial Radio Access Network (E-UTRAN) Node B (also commonly denoted as evolved Node B, enhanced Node B, eNodeB, or eNB) . One example of an NG-RAN base station is a next generation Node B (also sometimes referred to as a g Node B or gNB) .
[0006] A RAN provides its communication services with external entities through its connection to a core network (CN) . For example, E-UTRAN may utilize an Evolved Packet Core (EPC) while NG-RAN may utilize a 5G Core Network (5GC) .
[0007] BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS
[0008] To easily identify the discussion of any particular element or act, the most significant digit or digits in a reference number refer to the figure number in which that element is first introduced.
[0009] FIG. 1 illustrates an example of a high level structure for the use of an AI / ML model for time-spatial-frequency domain channel state information (CSI) compression, according to embodiments herein.
[0010] FIG. 2 illustrates an example flow of a procedure designed to enable time-spatial-frequency domain AI / ML based CSI feedback, according to embodiments herein.
[0011] FIG. 3 illustrates an example flow of a procedure design to enable time-spatial-frequency domain AI / ML based CSI feedback including triggering and resetting, according to embodiments herein.
[0012] FIG. 4 illustrates an example flow of a procedure design to enable time-spatial-frequency domain AI / ML based CSI feedback that accounts for cases where CSI feedback fails, according to embodiments herein.
[0013] FIG. 5 illustrates an example of indexing CSI reports, according to embodiments herein.
[0014] FIG. 6 illustrates an example of CSI report retransmission in cases of failed CSI report transmission, according to embodiments herein.
[0015] FIG. 7 illustrates an example of UCI mapping in cases of collision between a CSI report transmission and a CSI report retransmission, according to embodiments herein.
[0016] FIG. 8 illustrates a method of a UE, according to embodiments herein.
[0017] FIG. 9 illustrates a method of a base station, according to embodiments herein.
[0018] FIG. 10 illustrates an example architecture of a wireless communication system, according to embodiments disclosed herein.
[0019] FIG. 11 illustrates a system for performing signaling between a wireless device and a network device, according to embodiments disclosed herein.DETAILED DESCRIPTION
[0020] Various embodiments are described with regard to a UE. However, reference to a UE is merely provided for illustrative purposes. The example embodiments may be utilized with any electronic component that may establish a connection to a network and is configured with the hardware, software, and / or firmware to exchange information and data with the network. Therefore, the UE as described herein is used to represent any appropriate electronic component.
[0021] In some wireless communication systems, a channel state information (CSI) feedback enhancement may be considered. For example, for CSI compression cases where a two-sided artificial intelligence (AI) / machine learning (ML) model is used (where each of a UE and a base station implement at least a portion of the AI / ML model) , the trade-off between performance and complexity / overhead may be improved by extending the use of spatial / frequency compression into spatial / temporal / frequency compression.
[0022] Additional avenues for potential improvement to CSI compression cases include consideration of various cell / site specific models, and consideration of CSI compression plus prediction (as compared to a non-AI / ML based approach) . Additionally, issues related to inter-vendor training collaboration may be addressed. Still another avenue includes consideration of various performance gain metrics for CSI prediction (e.g., in a one-sided model) as compared to non-AI / ML based approaches (e.g., cell / site specific model could be considered to improve performance gain) .
[0023] Additionally, embodiments disclosed herein relate to model identifiers (IDs) that may be used to facilitate data collection, CSI compression and inferencing with respect to different use cases (e.g., different network-side conditions for the corresponding network signaling) . Within such contexts, it may be understood that a same model ID that is used in a radio resource control (RRC) configuration for a data collection procedure for training an AI / ML model is also used in a corresponding configuration for an inferencing procedure that uses the AI / ML model, in order to ensure compatibility / consistency as between the data collection procedure, CSI compression and the inferencing procedure. When so used, the model ID may thus be understood to indicate or represent, in an abstracted way, the applicable network-side condition for the corresponding network signaling.
[0024] Various options for the particularly identified items by a model ID may be considered. In a first option, the model ID identifies a CSI reconstruction mechanism that the network will use. In a second option, the model ID identifies a CSI generation mechanism that the UE will use. In a third option, the model ID identifies a paired CSI generation mechanism and CSI reconstruction mechanism. In a fourth option, the model ID represents a dataset ID used during type 3 sequential training. In a fifth option, the model ID represents a training session ID of a prior training session (e.g., based on an API) between the network and the UE. In a sixth option, the representative use of the model ID is designed / defined according to an offline co-engineering alignment, which transparent to a specification for the wireless communication system.
[0025] Note that in the event where a model ID most directly identifies one “side” (e.g., a UE side or network side) of an AI / ML model, the other corresponding device may be capable of determining its corresponding “side” of the overall AI / ML model using that model ID. Note also that in some embodiments, a model ID for CSI compression can be offline identified between different vendors.
[0026] Within some contexts applicable to various embodiments disclosed herein, the uses of a "dataset ID” versus a “model ID" may be understood as interchangeable. This may correspond to cases where, for example, a single dataset identified by the dataset ID is used to train a corresponding model identified by the model ID. In such a situation, each of the dataset ID and the model ID are logical IDs that correspond to the same scenario (e.g., the same network-side condition) , and thus a straight swap between a use of a dataset ID instead of using the corresponding model ID (or vice-versa) is possible.
[0027] FIG. 1 illustrates an example of a high level structure for the use of an AI / ML model for time-spatial-frequency domain CSI compression, according to embodiments herein.
[0028] In some mechanisms, AI / ML models used for CSI compression may account for frequency domain and spatial domain aspects. However, such frequency-spatial domain AI / ML models do not consider time domain aspects.
[0029] Embodiments herein relate to the use of time-spatial-frequency domain AI / ML models (which do consider, among other things, time domain aspects) . FIG. 1 accordingly illustrates that such an AI / ML model for CSI compression is made up of both an encoder 104 for encoding one or more elements of CSI feedback at the UE side prior to transmission and a decoder 106 for decoding the one or more elements of CSI feedback as encoded and as received the base station side.
[0030] Additionally, to account for time domain aspects, the use of an internal state at each of the encoder 104 and the decoder 106 corresponding to a given time may be introduced. For example, the encoder 104 (at a UE) , at a first time 102 (denoted t1) , uses a first eigen-vector 108 V1_t1 (e.g., of layer i of a precoding matrix indicator (PMI) of the CSI feedback) as an input to generate the first encoded CSI feedback 110, which is then transmitted by the UE, as illustrated. At this first time 102, the encoder 104 is in a first encoder state 114 (denoted S1enc) .
[0031] Further, as shown, corresponding to the first time 102, a decoder 106 (at a base station) receives the first encoded CSI feedback 110 and decodes it into a first reconstructed eigen-vector 112 (denoted V'1_t1) , which may then be used by the base station for purposes of precoder selection. At this first time 102, the decoder 106 is in a first decoder state 116 (denoted S1dec) .
[0032] Then, as illustrated, at a second time 118 (denoted t2) , the encoder 104 uses a second eigen-vector 120 (denoted V1_t2) (e.g., of the same layer i of the PMI of the CSI feedback) and further uses the first encoder state 114 (information about the encoder 104 as it was in the first encoder state 114 at the first time 102) to generate the second encoded CSI feedback 122, which is then transmitted by the UE, as illustrated. The use of the first encoder state 114 at this stage (and its corresponding effect on the generation of second encoded CSI feedback 122 at the encoder 104) represents an inclusion of time domain aspects at the encoder 104 of the time-spatial-frequency domain AI / ML model. At this second time 118, the encoder 104 is in a second encoder state 126 (denoted S2enc) .
[0033] Further, as shown, corresponding to the second time 118, the decoder 106 receives the second encoded CSI feedback 122 and decodes it into a second reconstructed eigen-vector 124 (denoted V'1_t2) , which may then be used by the base station for precoder selection. At this second time 118, the decoder 106 is in a second decoder state 128 (denoted S2dec) . During this decoding process, the decoder 106 takes into account the first decoder state 116 (information about the decoder 106 as it was in the first decoder state 116 at the first time 102) . The use of the first decoder state 116 at this stage (and its corresponding effect on the generation of second reconstructed eigen-vector 124 at the decoder 106) represents an inclusion of time domain aspects at the decoder 106 of the time-spatial-frequency domain AI / ML model.
[0034] As illustrated, at a third time 130 (denoted t3) , the encoder 104 uses a third eigen-vector 132 (denoted V1_t3) (e.g., of the same layer i of the PMI of the CSI feedback) and further uses the second encoder state 126 (information about the encoder 104 as it was in the second encoder state 126 at the second time 118) to generate the third encoded CSI feedback 134, which is then transmitted by the UE, as illustrated. The use of the second encoder state 126 at this stage (and its corresponding effect on the generation of third encoded CSI feedback 134 at the encoder 104) represents an inclusion of time domain aspects at the encoder 104 of the time-spatial-frequency domain AI / ML model. At this third time 130, the encoder 104 is in a third encoder state 138 (denoted S3enc) .
[0035] As shown, corresponding to the third time 130, the decoder 106 receives the third encoded CSI feedback 134 and decodes it into a third reconstructed eigen-vector 136 (denoted V'1_t3) , which may then be used by the base station for precoder selection. At this third time 130, the decoder 106 is in a third decoder state 140 (denoted S3dec) . During this decoding process, the decoder 106 takes into account the second decoder state 128 (information about the decoder 106 as it was in the second decoder state 128 at the second time 118) . The use of the second decoder state 128 at this stage (and its corresponding effect on the generation of third reconstructed eigen-vector 136 at the decoder 106) represents an inclusion of time domain aspects at the decoder 106 of the time-spatial-frequency domain AI / ML model.
[0036] Note that while FIG. 1 explicitly illustrates the use of a single layer i of a PMI, this is given by way of example only. The encoder 104 and the decoder 106 will be understood to be extendible to analogously encode / decode CSI feedback corresponding to multiple layers of a PMI (up to and including all layers of a PMI in some embodiments) .
[0037] In various embodiments, the encoder state information (e.g., the first encoder state 114, the second encoder state 126, and / or the third encoder state 138) may include or represent or correspond to one or more prior PMIs (e.g., prior to the corresponding time for that state) acquired by and encoded by (at least in part) the UE.
[0038] In various embodiments, the decoder state information (e.g., the first decoder state 116, the second decoder state 128, and / or the third decoder state 140) may include or represent or correspond to one or more prior PMIs (e.g., prior to the corresponding time for that state) that were previously decoded (at least in part) at the base station.
[0039] In some embodiments, each of the encoder 104 and / or the decoder 106 may be implemented using a recursive neural network (RNN) (e.g., a long short term memory (LTSM) RNN) . The information about the one or more prior PMIs (e.g., from the perspective of either the UE or the base station) may thus correspondingly be represented in the encoder 104 / the decoder 106 as weighting values within the corresponding RNN (and which may be modified as the encoder 104 / the decoder 106 is continuously operated with new inputs through time) .
[0040] Embodiments herein discuss signaling and procedure design to enable AI / ML based CSI feedback that reflects time domain, spatial domain, and frequency domain (collectively referred to herein as the time-spatial-frequency domain) aspects. For example, signaling and design aspects discussed herein may include an RRC configuration for a CSI report to enable time-spatial-frequency domain AI / ML based CSI feedback, triggering and resetting of time-spatial-frequency domain AI / ML based CSI feedback, a CSI feedback format for time-spatial-frequency domain AI / ML based CSI feedback, and / or one or more CSI feedback priority and / or omission rule (s) for time-spatial-frequency domain AI / ML based CSI feedback.
[0041] FIG. 2 illustrates an example flow of a procedure designed to enable time-spatial-frequency domain AI / ML based CSI feedback, according to embodiments herein.
[0042] In some embodiments, a procedure design may be introduced to enable time-spatial-frequency domain AI / ML based CSI feedback. For example, the time-spatial-frequency domain AI / ML based CSI feedback procedure may begin with a UE 202 reporting its supported AI / ML model (s) using corresponding supported AI / ML model ID (s) 206 to the base station 204. In some examples, the supported AI / ML model ID (s) 206 may be reported in a UE capability report. In other examples, the supported AI / ML model ID (s) 206 may be reported in RRCReconfigurationComplete message (e.g., when the UE receives a handover (HO) command to handover to a new cell) .
[0043] Subsequently, the base station 204 may transmit, to the UE 202, an RRC configuration 208 for a CSI report. In some cases, a supported reportConfigType information element (IE) of the RRC configuration 208 may include a semiPersistent IE transmission on a physical uplink control channel (PUCCH) and / or a semiPersistent IE transmission on a physical uplink shared channel (PUSCH) .
[0044] Further, the RRC configuration 208 may include a csi_ReportConfig IE and a csi_ResourceConfig IE. The csi_ReportConfig IE specifies that the particular csi_ResourceConfig IE (e.g., from multiple such csi_ReportConfig IEs) is to be used for measurement. Then, the csi_ResourceConfig IE indicates what type of reference signal the measurement is to be transmitted on and what type of transmissions are to be performed (e.g., as between periodic, aperiodic, semi-persistent) .
[0045] In some cases, a reportQuantity IE of the RRC configuration 208 may include a report quantity for a PMI used in CSI feedback. Note that this reported PMI may be based on / affected by the PMI of prior CSI feedback was stored in memory corresponding to prior state information within the AI / ML model (at an encoder and / or a decoder) , as previously described. In one example, the encoded PMI may be sent alone within the CSI report. Accordingly, the PMI may be defined separately in the reportQuantity IE from other potential value types for a CSI report. In such cases, it may be that a rank indicator (RI) value is understood to be the same as in the previous report (as it is understood that the use of a given AI / ML model over time should be consistent with respect to a number of layers being processed by that AI / ML model) . Additionally, a channel quality indicator (CQI) can be sent separately from the CSI report having the PMI in these cases.
[0046] In a second example, the PMI (e.g., as encoded by an encoder of the AI / ML model) may be jointly coded with the channel state information reference signal resource indicator (CRI) , RI, CQI, and layer indicator (LI) within a CSI report. The jointly encoded CRI-RI-PMI-CQI may correspond to a current subframe. Alternatively, the PMI and CQI values with the CSI report may be used for the current subframe, where the RI is the same as the RI from the previous report (again, as it is understood that the use of a given AI / ML model over time should be consistent with respect to a number of layers being processed by that AI / ML model) .
[0047] It should be understood that, although the reportQuantity IE and its corresponding items (i.e., CRI, RI, CQI, LI) are transmitted as part of a CSI report to the UE 202 in uplink control information (UCI) , the output of the encoder discussed herein is all or part of an encoded PMI, based on previously encoded PMIs, and the CRI, CQI and RI are calculated separately from the AI / ML encoding and decoding and then placed (e.g., as separate entities) within the CSI report in the UCI. It should also be understood that the encoding / decoding of the CSI feedback may analogously correspond to all or fewer than all layers of a PMI (up to and including all layers of a PMI) . In cases where the encoder / decoder of the time-spatial-frequency domain-based AI / ML model is used to encode / decode fewer than all layers of a PMI into the encoded PMI, it may be that another AI / ML model is used to encode remaining layers of the PMI into the encoded PMI.
[0048] The RRC configuration 208 may include one or more of a model ID, an AI / ML encoder input size including the number of subband bands, antenna ports and / or an AI / ML encoder output size (e.g., in UCI bits) . In some instances, a multiple UCI bit configuration may be utilized.
[0049] Subsequently, the base station 204 may transmit a DCI message including a triggering indication or resetting indication 210 to the UE 202 for either DCI-based triggering of or DCI-based resetting of the AI / ML model, as these processes are described in further detail elsewhere herein.
[0050] Then, as shown, the base station 204 may perform a channel state information reference signal (CSI-RS) transmission 212 to the UE 202. The UE 202 acquires CSI information (including, e.g., a PMI) for a channel between the UE and the base station 204 by measuring the CSI-RS. The UE 202 may then formulate a CSI report using that CSI-RS information. This CSI report may include an encoded PMI that is generated using the acquired PMI using one or more AI / ML models, as described herein.
[0051] The UE 202 then transmits UCI 214 to the base station 204 that includes a CSI report having the encoded PMI.
[0052] As shown, over time, subsequent, CSI-RS transmissions 216 and responsive transmissions of UCIs 218 may follow suit (e.g., until a further triggering indication or resetting indication 210 indication is transmitted as further detailed elsewhere herein) . These subsequent sets of CSI-RS transmission 216 and corresponding UCI 218 may use the same time-spatial-frequency AI / ML model for encoding / decoding PMI and thus may be affected by one or more prior PMIs, in the manner discussed herein.
[0053] FIG. 3 illustrates an example flow of a procedure design to enable time-spatial-frequency domain AI / ML based CSI feedback including triggering and resetting, according to embodiments herein.
[0054] As discussed herein (with reference to FIG. 2) , a procedure design to enable time-spatial-frequency domain AI / ML based CSI feedback begins with the UE 302 reporting its supported AI / ML model ID 306 to the base station 304 in, for example, a UE capability message. Subsequently, the base station 304 may transmit, to the UE 302, an RRC configuration 308 for a CSI report in the manner discussed in relation to FIG. 2.
[0055] The base station 304 may then transmit a triggering indication 310 to the UE 302. The base station 304 may use the triggering indication 310 to trigger a semi-persistent CSI reporting by the UE. Further, the base station 304 may later transmit a resetting indication 320 to the UE 302. The resetting indication 320 may be used to reset the semi-persistent CSI reporting as used by the UE. Various embodiments corresponding to these actions are now described.
[0056] In some embodiments, the triggering indication 310 is transmitted in a medium access control (MAC) control element (CE) (MAC CE) . Such cases may correspond to the triggering of semi-persistent CSI reporting on a PUCCH. It should be understood that the MAC CE triggering indication 310 may be used to start and / or restart CSI feedback that has been stopped in instances of low traffic where CSI feedback may not be needed. In some instances, the MAC CE triggering indication 310 may indicate the start of a time series of the semi-persistent CSI reporting. Corresponding to some such cases, it may be that no previous state information is used in either the UE side encoder and / or the network side decoder as the time series of CSI-RS measurements has been indicated to start anew.
[0057] In some other instances, the MAC CE may be sent to stop the transmission of CSI feedback. For example, when the network does not require a PMI (such as when there is no traffic to transmit) and the base station 304 may provide the UE 302 with a MAC CE that indicates that the UE may stop the semi-persistent CSI feedback transmissions. In cases where a MAC CE stops the semi-persistent CSI feedback transmissions, and at a later time triggers more semi-persistent CSI feedback transmission (s) , the AI / ML model used to encode / decode the PMI of the CSI feedback may started from an initial state (e.g., the encoder may be started from an initial encoder state and the decoder may be started from an initial decoder state) .
[0058] In yet some other instances, a MAC CE may reset the use of semi-persistent CSI feedback (i.e., the base station 304 may transmit a resetting indication 320 indication to the UE 302, as illustrated) . In this case, the AI / ML model used to encode / decode PMI of the CSI feedback starts from the initial state again. The resetting of the semi-persistent CSI feedback may occur when a large amount of dropped CSI feedback is observed and the encoder state and decoder state are mismatched at the AI / ML model at the UE 302 and at the base station 304 accordingly. Such resetting of the semi-persistent CSI feedback can be understood to act as resetting the memory of the encoder and / or decoder of the AI / ML model discussed herein whereby the encoder / decoder is understood to have returned to the corresponding initial state (thus re-synching the state of the encoder and decoder of the AI / ML model) .
[0059] In some embodiments, the triggering indication 310 is transmitted in a DCI. Such cases may correspond to the triggering of semi-persistent CSI reporting on a PUSCH. In such cases, bits may be used in the DCI for the triggering indication 310.
[0060] In some embodiments, the resetting indication 320 is transmitted in a DCI. In such cases, new bits may be used in the DCI for the resetting indication 320.
[0061] The semi-persistent CSI reporting may use various DCI formats, including DCI format 0_1, 0_2 and 0_3. In some examples, a CSI request bit may be used to perform the DCI triggering. Correspondingly, one state in a CSI report trigger list may indicate to start and / or reset the CSI feedback. Further, one state in the CSI report trigger list may indicate the end of the CSI feedback.
[0062] In some examples, new bit (s) in the DCI may be used to indicate the start, reset, and / or end of the AI / ML model based semi-persistent CSI feedback.
[0063] It should be understood that, after the triggering indication 310 is transmitted, the base station 304 may perform a CSI-RS transmission 312 in a channel between the UE 302 and the base station 304 to the UE 302 and in response, the UE may generate a corresponding CSI report (including an encoded PMI generated at least in part using an encoder of an AI / ML model) that is then transmitted in the UCI 314. The base station 304 receives the UCI and uses a decoder of the AI / ML model to decode at least part of a reconstructed PMI using the encoded PMI received in the UCI 314. This process may be repeated (see, e.g., the CSI-RS transmission 316 and the UCI 318) according to the semi-persistent configuration for the CSI feedback until a resetting indication 320 is triggered (e.g., for reasons described herein)
[0064] After a resetting indication 320 indication is received by the UE 302, any state information used by the encoder / decoder of the AI / ML model is dropped and the encoder and the decoder of the AI / ML model are thus returned to an initial state. The semi-persistent CSI process may then be continued (see, e.g., the CSI-RS transmission 322 and the UCI 324) according to the semi-persistent configuration for the CSI feedback in a manner that is affected by prior PMI information generated prior to the resetting indication 320. In some cases, this is used to re-coordinate states at each of the encoder and the decoder of the AI / ML model when these have become out of synchronization.
[0065] In some mechanisms, CSI feedback contains two parts. The first part (i.e., Part 1) is a fixed size and may indicate the size of the second part (i.e., Part 2) . Accordingly, the second part has a dynamic / flexible size. The first part may include a RI, CQI and the size of the second part. The second part may include the parameters for a codebook based PMI. For example, the PMI payload size depends on the rank UE selected, and the non-zero coefficients based on a parameter configuration.
[0066] However, in certain embodiments, for time-spatial-frequency domain AI / ML based CSI feedback, the AI / ML model may be a layer by layer model, therefore, for each feedback, a similar number of layers may be needed. Accordingly, in some cases, the RI is the same for each report because the RI is a long term statistic (i.e., it may not change during the semi-persistent CSI feedback, until it is reset) . The layer output is specified in a CSI-reportConfig IE, therefore the total overhead may be fixed. In some such examples, the information may be in the first part (i.e., in Part 1) .
[0067] In some cases, an RI may be the same or different for each report. The encoder of the time-spatial-frequency domain AI / ML model may encode layer 1 and / or layer 2, and for layer 3 and layer 4, a separate encoder of the AI / ML model that encodes only according to spatial-frequency domain aspects (and not according to time domain aspects) is used. In some such examples, the first part (Part 1) of a CSI report may indicate the RI and the second part of the CSI report (Part 2) may include a CSI report for layer 3 and layer 4. The CSI report for layer 1 and layer 2 may be included either in the first part or the second part, depending on embodiment.
[0068] In some embodiments for CSI feedback, in one example, different bits have different priorities. The spatial basis index may have a higher priority. For non-zero coefficients, a specific order may be designed. In such examples, when a UCI container is not large enough for all information, the lower priority bits are dropped.
[0069] Embodiments herein for time-spatial-frequency domain AI / ML based CSI feedback contemplate cases where the time-spatial-frequency domain CSI feedback has a higher priority as compared to the spatial-frequency domain CSI feedback.
[0070] In some instances, mechanisms provide a CSI priority rule. For example, CSI reports may be associated with the priority value PriiCSI (y, k, c, s) = 2 ·Ncells ·Ms ·y + Ncells ·Ms ·k + Ms ·c + s, where y = 0 when aperiodic CSI reports are to be carried on a PUSCH, y = 1 when semi-persistent CSI reports are to be carried on a PUSCH, y = 2 when semi-persistent CSI reports are to be carried on a PUCCH, and y = 3 when periodic CSI reports are to be carried on a PUCCH. Additionally, k = 0 for CSI reports carrying layer 1 reference signal received power (L1-RSRP) or a layer 1 signal to noise and interference (L1-SINR) , and k = 1 for CSI reports not carrying L1-RSRP or L1-SINR; c is the serving cell index; Ncells is the value of the higher layer parameter maxNrofServingCells IE; s is the reportConfigID IE; and Ms is the value of the higher layer parameter maxNrofCSI-ReportConfigurations IE. A first CSI report may have priority over a second CSI report if the associated PriiCSI (y, k, c, s) value is lower for the first CSI report than for the second CSI report. When the two reports collide, the lower priority report may be dropped.
[0071] In certain embodiments proposed for CSI reports for time-spatial-frequency AI / ML based CSI feedback, it is proposed that a corresponding value of k, discussed herein, is set equal to 0, such that the report has the same priority as a report that carries L1-RSRP and / or an L1-SINR, and has a higher priority as compared to other CSI reports. In other embodiments, it may be that k, discussed herein, is set equal to 1, the such that such a CSI report has the same priority as a CSI report not carrying an L1-RSRP or an L1-SINR.
[0072] As is described in various embodiments herein for time-spatial-frequency domain AI / ML based CSI feedback, a base station / network state as used by a decoder of an AI / ML model is updated based on each UE feedback transmission generated by the corresponding encoder of the AI / ML model. Accordingly, when UE feedback is dropped, received incorrectly, or collides with a higher priority transmission, the state of the decoder at the base station / network may become mismatched with the state of the encoder at the UE. For example, when a large number of UCI transmissions having the corresponding CSI feedback are not received correctly, this state mismatch may be large, thus causing performance loss.
[0073] In instances of CSI feedback that does not account for the time domain, if PMI is dropped or lost, the base station / network may simply be configured use a previous PMI until a new PMI is transmitted and correctly received, thus causing inaccuracy and performance loss. Therefore, there is no CSI retransmission and no forward error propagation issue.
[0074] However, as various embodiments herein relate to the use of time-spatial-frequency domain AI / ML based CSI feedback that does account for time domain aspects, the outright dropping of a missed PMI may create forward error propagation issues. Accordingly, embodiments herein discuss cases of the retransmission of UCI (including CSI feedback) that may be used to preserve alignment of the state of the decoder at the base station and the state of the encoder at the UE for time-spatial-frequency domain AI / ML based CSI feedback.
[0075] FIG. 4 illustrates an example flow of a procedure design to enable time-spatial-frequency domain AI / ML based CSI feedback that accounts for cases where CSI feedback fails, according to embodiments herein.
[0076] In some embodiments, the base station 404 may trigger a semi-persistent UE report. As discussed elsewhere herein (e.g., in FIG. 2 and FIG. 3 and related discussion) , the base station 404 may transmit, to the UE 402, an RRC configuration 406 and a triggering indication 408 triggering semi-persistent CSI reporting according to the RRC configuration 406. A CSI-RS transmission 410 (corresponding to State 0) is accordingly transmitted. In response, the UE 402 may transmit a semi-persistent UCI 412 back to the base station 404 on, for example, a PUSCH or a PUCCH.
[0077] Accordingly, the base station 404 may perform a cyclic redundancy check (CRC) on the UCI 412. The CRC uses corresponding bits that may be attached to the UCI 412 if the payload size of the UCI 412 is greater than or equal to a given number of bits (e.g., 11 bits) In some instances, if the CRC does not fail, the base station 404 may proceed normally, eventually transmitting a second CSI-RS transmission 414 to the UE 402 for State 1 (as illustrated) .
[0078] Similarly to the case for the State 0, for State 1, the UE 402 may transmit a UCI 416 to the base station 404, and the base station 404 checks the CRC of the UCI 416. However, in this case, the CRC fails, and the base station 404 accordingly concludes that the UCI 416 may be wrong. In such cases where the CRC fails and the UCI 416 may be wrong, for the precoding matrix calculation that is performed before a next UCI report arrives, the base station 404 may temporally use the previous decoder output for various corresponding calculations.
[0079] Further, to prevent forward error propagation issues, the base station 404 may trigger an aperiodic CSI feedback for CSI feedback retransmission, using, for example, DCI. The trigger may indicate, to the UE 402, that the UCI 416 corresponding to the previous state (State 1) should be retransmitted to the base station 404. For example, the base station 404 may use a DCI to send a retransmission indication 418 for corresponding to the UCI 416 and the UE 402, in response, may retransmit a the UCI 416 that was identified as incorrect. In the example illustrated in FIG. 4, the retransmission indication 418 and the responsive retransmission of the UCI 416 may correspond to the previously failed initial transmission UCI 416 at the State 1 stage. In this way, the base station 404 is informed of a version of the UCI 416 that is correct, and thus uses the encoded PMI therein for base station decoder state determination purposes. In this way, forward error propagation may be avoided.
[0080] Additional CSI-RS transmissions 420 and corresponding UCIs 422 discussed herein may then be performed for each State up to State N where the CRC of each UCI report is checked / verified for success or failure (e.g., until reset of the State is indicated as discussed herein in FIG. 3) .
[0081] FIG. 5 illustrates an example of indexing CSI reports, according to embodiments herein. FIG. 5 illustrates a first CSI report 502, a second CSI repot 504, a third CSI report 506, a fourth CSI report 508, a fifth CSI report 510, a sixth CSI report 512, a seventh CSI report 514, and an eighth CSI report 516.
[0082] In certain embodiments, in order to enable retransmission of the CSI reports, the CSI reports may be indexed. For example, a first CSI report 502 may have an index of zero, a second CSI repot 504 may have an index of one, a third CSI report 506 may have an index of two, and so on, as illustrated.
[0083] The CSI feedback index may take up a configured amount of bits (e.g., two bits, three bits, or four bits) and may be periodically / circularly reused, as also illustrated. For instance, if the index takes up two bits (as in FIG. 5) , the first CSI report 502 and the fifth CSI report 510 may have an index of zero, as illustrated in FIG. 5.
[0084] It will be noted that due to index reuse, there may be ambiguity; for example, in the two-bit case illustrated in FIG. 5, there may be ambiguity after four CSI reports backward in time. Accordingly, the UE and / or the base station may be configured to understand that an index refers to the closest CSI report in time.
[0085] An index for a CSI report may be sent together with the CSI report (e.g., as part of CSI Part 1 or CSI Part 2) . In some instances, the size of the index is indicated by the base station to the UE by RRC configuration (i.e., the size of the index corresponds to the RRC configuration) .
[0086] FIG. 6 illustrates an example of CSI report retransmission in cases of failed CSI report transmission, according to embodiments herein.
[0087] In some instances, aperiodic CSI feedback may be triggered for a retransmission of a failed CSI report. For example, if a first CSI report 602 fails, a base station may transmit a first DCI trigger 604 for an aperiodic retransmission of the CSI report. Accordingly, a first CSI report retransmission 606 may occur between the failed first CSI report 602 and the subsequent second CSI report 608.
[0088] A similar procedure may occur where a second DCI trigger 612 for CSI feedback retransmission may be sent in response to a failure of a third CSI report 610. In response to the second DCI trigger 612, a UE may perform a third CSI report retransmission 614.
[0089] It is noted that, as discussed herein, that the retransmitted CSI report is used to update the state information of the decoder at the base station so that there is no mismatch between the state of the encoder and the state of the decoder at the UE in future instances of time-spatial-frequency domain AI / ML encoding and decoding. In cases where the retransmitted CSI report is received before the time for the next CSI report according to the semi-persistent CSI reporting configuration, it may be incorporated into the state information of the decoder at the network (and thus used for decoding and subsequently precoder selection) corresponding to that next CSI report.
[0090] In some cases, the DCI trigger indication may take the form of a DCI including, for example, DCI format 0-1, DCI format 0-2 and DCI format 0-3. A CSI report retransmission request / DCI trigger may indicate the corresponding semi-persistent CSI report configuration. Additional bits to indicate an applicable CSI report index for the CSI report being requested may be included in the DCI trigger.
[0091] FIG. 7 illustrates an example of UCI mapping in cases of collision between a CSI report transmission and a CSI report retransmission, according to embodiments herein.
[0092] In certain embodiments, if a retransmitted CSI report does not collide with a semi-persistent CSI report transmission (as discussed in FIG. 6) (e.g., assuming the case of aperiodic CSI reporting on a PUSCH) , the retransmitted CSI report may be mapped to a UCI bit sequence that corresponds to a normal CSI report bit sequence map.
[0093] In some examples, the base station may avoid scheduling retransmission of a CSI report retransmission such that it collides with a CSI report being sent to the semi-persistent CSI feedback configuration.
[0094] In some embodiments, an offset 706 between a DCI trigger 704 transmitted after a failed first CSI report 702 and the first CSI report retransmission 708 may be used. In some cases, the offset 706 is set / reduced as to avoid collision between the first CSI report retransmission 708 and the next second CSI report 710 (note that FIG. 7 illustrates the alternative collision case) .
[0095] In some embodiments if the first CSI report retransmission 708 collides with second CSI report 710 that is being sent according to the semi-persistent CSI feedback configuration (as shown in FIG. 7) the first CSI report retransmission 708 may have a lower priority as compared to the second CSI report 710. The CSI field mapping order may take the form of prioritizing the semi-persistent second CSI report 710 first, and the first CSI report retransmission 708 second. Further, if the resulting payload size is too big for UCI of the UE, the first CSI report retransmission 708 may be dropped. In cases where the first CSI report retransmission 606 fails or is dropped as discussed, the network may trigger another first CSI report retransmission 708 according to examples discussed herein.
[0096] In certain cases, for CSI reporting, a trigger state in DCI may be linked with multiple CSI reports. For example, a trigger state may be linked to at least one retransmitted CSI report, and at least one non-retransmitted CSI report. In some cases, the retransmitted CSI report (s) may have a higher priority than the non-retransmitted CSI report (s) . Additionally, among the retransmitted CSI reports, a priority rule may apply. For example, a list of priorities from highest priority to lowest priority may be as follows: a retransmitted CSI report-1 priority > retransmitted CSI report-2 priority > non-retransmitted CSI report-1 priority > non-retransmitted CSI report-2 priority.
[0097] FIG. 8 illustrates a method 800 of a UE, according to embodiments herein. The illustrated method 800 includes receiving 802, from a base station, a CSI reporting configuration that identifies an AI / ML model and that configures the UE to perform CSI reporting using the AI / ML model. The method 800 further includes acquiring 804 a PMI for a channel between the UE and the base station using a CSI-RS transmitted by the base station. The method 800 further includes encoding 806, at the UE, an encoded PMI by applying at least a first portion of the PMI with a first encoder of the AI / ML model, wherein the first encoder of the AI / ML model uses information corresponding to one or more prior PMIs acquired by the UE to perform encoding for the encoded PMI. The method 800 further includes transmitting 808, to the base station, UCI comprising a first CSI report of the CSI reporting, the first CSI report comprising the encoded PMI.
[0098] In some embodiments of the method 800, the first encoder of the AI / ML model comprises an RNN that is used to encode the encoded PMI and that stores the information corresponding to the one or more prior PMIs as weighting values.
[0099] In some embodiments of the method 800, the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting on one of a PUCCH and a PUSCH as part of the CSI reporting.
[0100] In some embodiments of the method 800, the CSI report configuration further configures the UE for CQI use in the CSI reporting, and wherein the first CSI report further comprises a CQI acquired by the UE using the CSI-RS.
[0101] In some embodiments of the method 800, the CSI report configuration further configures the UE for RI use in the CSI reporting, and wherein the first CSI report further comprises a RI that is equal to a prior RI reported corresponding to a prior use of the AI / ML model.
[0102] In some embodiments, the method 800 further comprises transmitting, to the base station, capability information indicating that the UE can use the AI / ML model. In some such embodiments, the capability information is sent to the base station as one of: a UE capability report message; and an RRC reconfiguration complete message that is sent as part of a handover of the UE to the base station.
[0103] In some embodiments, the method 800 further comprises receiving, from the base station, a MAC CE that triggers the UE to activate the CSI reporting.
[0104] In some embodiments, the method 800 further comprises receiving, from the base station, a MAC CE that resets the first encoder to an encoder initial state.
[0105] In some embodiments, the method 800 further comprises receiving, from the base station, DCI that triggers the UE to activate the CSI reporting.
[0106] In some embodiments, the method 800 further comprises receiving, from the base station, DCI that resets the first encoder to an encoder initial state.
[0107] In some embodiments of the method 800, the UE further encodes the encoded PMI by applying a second portion of the PMI with a second encoder of the AI / ML model that does not use the information corresponding to the one or more prior PMIs.
[0108] In some embodiments, the method 800 further comprises determining that a UCI container size used by the UE is not large enough to contain both the first CSI report and a second CSI report that does not include L1-RSRP information or L1-SINR information, and dropping the second CSI report.
[0109] In some embodiments, the method 800 further comprises receiving, from the base station, a retransmission trigger for the first CSI report, and retransmitting, to the base station, the first CSI report. In some such embodiments, the first CSI report corresponds to an index according to the CSI reporting configuration, the retransmission trigger identifies the CSI reporting configuration and the index corresponding to the first CSI report, and the UE identifies the first CSI report for retransmission using the index. In some other such embodiments, the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting as part of the CSI reporting, and wherein UE retransmits the first CSI report separately from the semi-persistent CSI reporting.
[0110] In some embodiments, the method 800 further comprises receiving, from the base station, a retransmission trigger for the first CSI report, determining that a UCI container size used by the UE is not large enough to contain both a retransmission of the first CSI report and a second CSI report of the CSI reporting, and dropping the retransmission of the first CSI report.
[0111] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein) .
[0112] Embodiments contemplated herein include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform one or more elements of the method 800. This non-transitory computer-readable media may be, for example, a memory of a UE (such as a memory 1106 of a wireless device 1102 that is a UE, as described herein) .
[0113] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein) .
[0114] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 800. This apparatus may be, for example, an apparatus of a UE (such as a wireless device 1102 that is a UE, as described herein) .
[0115] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 800.
[0116] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by a processor is to cause the processor to carry out one or more elements of the method 800. The processor may be a processor of a UE (such as a processor (s) 1104 of a wireless device 1102 that is a UE, as described herein) . These instructions may be, for example, located in the processor and / or on a memory of the UE (such as a memory 1106 of a wireless device 1102 that is a UE, as described herein) .
[0117] FIG. 9 illustrates a method 900 of a base station, according to embodiments herein. The illustrated method 900 includes sending 902, to a UE, a CSI reporting configuration that identifies an AI / ML model and that configures the UE to perform CSI reporting using the AI / ML model. The method 900 further includes transmitting 904, to the UE, a CSI-RS. The method 900 further includes receiving 906, from the UE, UCI comprising a first CSI report of the CSI reporting, the first CSI report comprising a first encoded PMI corresponding to the CSI-RS. The method 900 further includes decoding 908, at the base station, a PMI from the encoded PMI by applying at least a first portion of the encoded PMI with a first decoder of the AI / ML model, wherein the first decoder of the AI / ML model uses information corresponding to one or more prior PMIs decoded by the base station to perform decoding for the PMI. The method 900 further includes selecting 910 a precoder for use with the UE based on the PMI.
[0118] In some embodiments of the method 900, the first decoder comprises an RNN that is used to decode the encoded PMI and that stores the information corresponding to the one or more prior PMIs as weighting values.
[0119] In some embodiments of the method 900, the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting on one of a PUCCH and a PUSCH as part of the CSI reporting.
[0120] In some embodiments of the method 900, the CSI report configuration further configures the UE for CQI use in the CSI reporting, and wherein the first CSI report further comprises a CQI acquired by the UE using the CSI-RS.
[0121] In some embodiments of the method 900, the CSI report configuration further configures the UE for RI use in the CSI reporting, and wherein the first CSI report further comprises a RI that is equal to a prior RI reported corresponding to a prior use of the AI / ML model.
[0122] In some embodiments, the method 900 further comprises receiving, from the UE, capability information indicating that the UE can use the AI / ML model. In some such embodiments, the capability information is received from the UE as one of: a UE capability report message, and an RRC reconfiguration complete message that is sent as part of a handover of the UE to the base station.
[0123] In some embodiments, the method 900 further comprises sending, to the UE, a MAC CE that triggers the UE to activate the CSI reporting.
[0124] In some embodiments, the method 900 further comprises determining that the first decoder is out of sync with an encoder of the AI / ML model used by the UE to generate the encoded PMI, sending, to the UE, a MAC CE that instructs the UE to reset the encoder to an encoder initial state, and resetting the first decoder to a decoder initial state.
[0125] In some embodiments, the method 900 further comprises sending, to the UE, DCI that triggers the UE to activate the CSI reporting.
[0126] In some embodiments, the method 900 further comprises determining that the first decoder is out of sync with an encoder of the AI / ML model used by the UE to generate the encoded PMI, sending, to the UE, DCI that instructs the UE to reset the encoder to an encoder initial state, and resetting the first decoder to a decoder initial state.
[0127] In some embodiments of the method 900, the base station further decodes the PMI by applying a second portion of the encoded PMI with a second decoder of the AI / ML model that does not use the information corresponding to the one or more prior PMIs.
[0128] In some embodiments, the method 900 further comprises determining to instruct the UE to retransmit the first CSI report, sending, to the UE, a retransmission trigger for the first CSI report, and receiving, from the UE, a retransmission of the first CSI report. In some such embodiments, the base station determines to instruct the UE to retransmit the first CSI report based on a CRC of the UCI. In some other such embodiments, the first CSI report corresponds to an index according to the CSI reporting configuration, and the retransmission trigger identifies the CSI reporting configuration and the index corresponding to the first CSI report. In yet some other such embodiments, the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting as part of the CSI reporting, and wherein the retransmission of the first CSI report is received separately from the semi-persistent CSI reporting.
[0129] Embodiments contemplated herein include an apparatus comprising means to perform one or more elements of the method 900. This apparatus may be, for example, an apparatus of a base station (such as a network device 1118 that is a base station, as described herein) .
[0130] Embodiments contemplated herein include one or more non-transitory computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform one or more elements of the method 900. This non-transitory computer-readable media may be, for example, a memory of a base station (such as a memory 1122 of a network device 1118 that is a base station, as described herein) .
[0131] Embodiments contemplated herein include an apparatus comprising logic, modules, or circuitry to perform one or more elements of the method 900. This apparatus may be, for example, an apparatus of a base station (such as a network device 1118 that is a base station, as described herein) .
[0132] Embodiments contemplated herein include an apparatus comprising: one or more processors and one or more computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more elements of the method 900. This apparatus may be, for example, an apparatus of a base station (such as a network device 1118 that is a base station, as described herein) .
[0133] Embodiments contemplated herein include a signal as described in or related to one or more elements of the method 900.
[0134] Embodiments contemplated herein include a computer program or computer program product comprising instructions, wherein execution of the program by a processing element is to cause the processing element to carry out one or more elements of the method 900. The processor may be a processor of a base station (such as a processor (s) 1120 of a network device 1118 that is a base station, as described herein) . These instructions may be, for example, located in the processor and / or on a memory of the base station (such as a memory 1122 of a network device 1118 that is a base station, as described herein) .
[0135] FIG. 10 illustrates an example architecture of a wireless communication system 1000, according to embodiments disclosed herein. The following description is provided for an example wireless communication system 1000 that operates in conjunction with the LTE system standards and / or 5G or NR system standards as provided by 3GPP technical specifications.
[0136] As shown by FIG. 10, the wireless communication system 1000 includes UE 1002 and UE 1004 (although any number of UEs may be used) . In this example, the UE 1002 and the UE 1004 are illustrated as smartphones (e.g., handheld touchscreen mobile computing devices connectable to one or more cellular networks) , but may also comprise any mobile or non-mobile computing device configured for wireless communication.
[0137] The UE 1002 and UE 1004 may be configured to communicatively couple with a RAN 1006. In embodiments, the RAN 1006 may be NG-RAN, E-UTRAN, etc. The UE 1002 and UE 1004 utilize connections (or channels) (shown as connection 1008 and connection 1010, respectively) with the RAN 1006, each of which comprises a physical communications interface. The RAN 1006 can include one or more base stations (such as base station 1012 and base station 1014) that enable the connection 1008 and connection 1010.
[0138] In this example, the connection 1008 and connection 1010 are air interfaces to enable such communicative coupling, and may be consistent with RAT (s) used by the RAN 1006, such as, for example, an LTE and / or NR.
[0139] In some embodiments, the UE 1002 and UE 1004 may also directly exchange communication data via a sidelink interface 1016. The UE 1004 is shown to be configured to access an access point (shown as AP 1018) via connection 1020. By way of example, the connection 1020 can comprise a local wireless connection, such as a connection consistent with any IEEE 802.11 protocol, wherein the AP 1018 may comprise a router. In this example, the AP 1018 may be connected to another network (for example, the Internet) without going through a CN 1024.
[0140] In embodiments, the UE 1002 and UE 1004 can be configured to communicate using orthogonal frequency division multiplexing (OFDM) communication signals with each other or with the base station 1012 and / or the base station 1014 over a multicarrier communication channel in accordance with various communication techniques, such as, but not limited to, an orthogonal frequency division multiple access (OFDMA) communication technique (e.g., for downlink communications) or a single carrier frequency division multiple access (SC-FDMA) communication technique (e.g., for uplink and ProSe or sidelink communications) , although the scope of the embodiments is not limited in this respect. The OFDM signals can comprise a plurality of orthogonal subcarriers.
[0141] In some embodiments, all or parts of the base station 1012 or base station 1014 may be implemented as one or more software entities running on server computers as part of a virtual network. In addition, or in other embodiments, the base station 1012 or base station 1014 may be configured to communicate with one another via interface 1022. In embodiments where the wireless communication system 1000 is an LTE system (e.g., when the CN 1024 is an EPC) , the interface 1022 may be an X2 interface. The X2 interface may be defined between two or more base stations (e.g., two or more eNBs and the like) that connect to an EPC, and / or between two eNBs connecting to the EPC. In embodiments where the wireless communication system 1000 is an NR system (e.g., when CN 1024 is a 5GC) , the interface 1022 may be an Xn interface. The Xn interface is defined between two or more base stations (e.g., two or more gNBs and the like) that connect to 5GC, between a base station 1012 (e.g., a gNB) connecting to 5GC and an eNB, and / or between two eNBs connecting to 5GC (e.g., CN 1024) .
[0142] The RAN 1006 is shown to be communicatively coupled to the CN 1024. The CN 1024 may comprise one or more network elements 1026, which are configured to offer various data and telecommunications services to customers / subscribers (e.g., users of UE 1002 and UE 1004) who are connected to the CN 1024 via the RAN 1006. The components of the CN 1024 may be implemented in one physical device or separate physical devices including components to read and execute instructions from a machine-readable or computer-readable medium (e.g., a non-transitory machine-readable storage medium) .
[0143] In embodiments, the CN 1024 may be an EPC, and the RAN 1006 may be connected with the CN 1024 via an S1 interface 1028. In embodiments, the S1 interface 1028 may be split into two parts, an S1 user plane (S1-U) interface, which carries traffic data between the base station 1012 or base station 1014 and a serving gateway (S-GW) , and the S1-MME interface, which is a signaling interface between the base station 1012 or base station 1014 and mobility management entities (MMEs) .
[0144] In embodiments, the CN 1024 may be a 5GC, and the RAN 1006 may be connected with the CN 1024 via an NG interface 1028. In embodiments, the NG interface 1028 may be split into two parts, an NG user plane (NG-U) interface, which carries traffic data between the base station 1012 or base station 1014 and a user plane function (UPF) , and the S1 control plane (NG-C) interface, which is a signaling interface between the base station 1012 or base station 1014 and access and mobility management functions (AMFs) .
[0145] Generally, an application server 1030 may be an element offering applications that use internet protocol (IP) bearer resources with the CN 1024 (e.g., packet switched data services) . The application server 1030 can also be configured to support one or more communication services (e.g., VoIP sessions, group communication sessions, etc. ) for the UE 1002 and UE 1004 via the CN 1024. The application server 1030 may communicate with the CN 1024 through an IP communications interface 1032.
[0146] FIG. 11 illustrates a system 1100 for performing signaling 1134 between a wireless device 1102 and a network device 1118, according to embodiments disclosed herein. The system 1100 may be a portion of a wireless communications system as herein described. The wireless device 1102 may be, for example, a UE of a wireless communication system. The network device 1118 may be, for example, a base station (e.g., an eNB or a gNB) of a wireless communication system.
[0147] The wireless device 1102 may include one or more processor (s) 1104. The processor (s) 1104 may execute instructions such that various operations of the wireless device 1102 are performed, as described herein. The processor (s) 1104 may include one or more baseband processors implemented using, for example, a central processing unit (CPU) , a digital signal processor (DSP) , an application specific integrated circuit (ASIC) , a controller, a field programmable gate array (FPGA) device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.
[0148] The wireless device 1102 may include a memory 1106. The memory 1106 may be a non-transitory computer-readable storage medium that stores instructions 1108 (which may include, for example, the instructions being executed by the processor (s) 1104) . The instructions 1108 may also be referred to as program code or a computer program. The memory 1106 may also store data used by, and results computed by, the processor (s) 1104.
[0149] The wireless device 1102 may include one or more transceiver (s) 1110 that may include radio frequency (RF) transmitter circuitry and / or receiver circuitry that use the antenna (s) 1112 of the wireless device 1102 to facilitate signaling (e.g., the signaling 1134) to and / or from the wireless device 1102 with other devices (e.g., the network device 1118) according to corresponding RATs.
[0150] The wireless device 1102 may include one or more antenna (s) 1112 (e.g., one, two, four, or more) . For embodiments with multiple antenna (s) 1112, the wireless device 1102 may leverage the spatial diversity of such multiple antenna (s) 1112 to send and / or receive multiple different data streams on the same time and frequency resources. This behavior may be referred to as, for example, multiple input multiple output (MIMO) behavior (referring to the multiple antennas used at each of a transmitting device and a receiving device that enable this aspect) . MIMO transmissions by the wireless device 1102 may be accomplished according to precoding (or digital beamforming) that is applied at the wireless device 1102 that multiplexes the data streams across the antenna (s) 1112 according to known or assumed channel characteristics such that each data stream is received with an appropriate signal strength relative to other streams and at a desired location in the spatial domain (e.g., the location of a receiver associated with that data stream) . Certain embodiments may use single user MIMO (SU-MIMO) methods (where the data streams are all directed to a single receiver) and / or multi user MIMO (MU-MIMO) methods (where individual data streams may be directed to individual (different) receivers in different locations in the spatial domain) .
[0151] In certain embodiments having multiple antennas, the wireless device 1102 may implement analog beamforming techniques, whereby phases of the signals sent by the antenna (s) 1112 are relatively adjusted such that the (joint) transmission of the antenna (s) 1112 can be directed (this is sometimes referred to as beam steering) .
[0152] The wireless device 1102 may include one or more interface (s) 1114. The interface (s) 1114 may be used to provide input to or output from the wireless device 1102. For example, a wireless device 1102 that is a UE may include interface (s) 1114 such as microphones, speakers, a touchscreen, buttons, and the like in order to allow for input and / or output to the UE by a user of the UE. Other interfaces of such a UE may be made up of transmitters, receivers, and other circuitry (e.g., other than the transceiver (s) 1110 / antenna (s) 1112 already described) that allow for communication between the UE and other devices and may operate according to known protocols (e.g., and the like) .
[0153] The wireless device 1102 may include a PMI encoding module 1116. The PMI encoding module 1116 may be implemented via hardware, software, or combinations thereof. For example, the PMI encoding module 1116 may be implemented as a processor, circuit, and / or instructions 1108 stored in the memory 1106 and executed by the processor (s) 1104. In some examples, the PMI encoding module 1116 may be integrated within the processor (s) 1104 and / or the transceiver (s) 1110. For example, the PMI encoding module 1116 may be implemented by a combination of software components (e.g., executed by a DSP or a general processor) and hardware components (e.g., logic gates and circuitry) within the processor (s) 1104 or the transceiver (s) 1110.
[0154] The PMI encoding module 1116 may be used for various aspects of the present disclosure, for example, aspects of any of FIG. 1, FIG. 2, FIG. 3, FIG. 4, FIG. 5, FIG. 6, FIG. 7, FIG. 8 and FIG. 10. The PMI encoding module 1116 is configured to receive a CSI reporting configuration that identifies an AI / ML model and configures the UE to perform CSI reporting using an AI / ML model. The PMI encoding module 1116 is further configured to acquire a PMI for a channel between the UE and the base station using a CSI-RS transmitted by the base station. Additionally, the PMI encoding module 1116 is configured to encode, at the UE, an encoded PMI by applying at least a first portion of the PMI with a first encoder of the AI / ML model, wherein the first encoder of the AI / ML model uses information corresponding prior PMIs acquired by the UE to perform encoding for the encoded PMI. The PMI encoding module 1116 is further configured to prepare the encoded PMI for transmission to the base station in UCI where the UCI includes a CSI report of the CSI reporting that includes the PMI.
[0155] The network device 1118 may include one or more processor (s) 1120. The processor (s) 1120 may execute instructions such that various operations of the network device 1118 are performed, as described herein. The processor (s) 1120 may include one or more baseband processors implemented using, for example, a CPU, a DSP, an ASIC, a controller, an FPGA device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein.
[0156] The network device 1118 may include a memory 1122. The memory 1122 may be a non-transitory computer-readable storage medium that stores instructions 1124 (which may include, for example, the instructions being executed by the processor (s) 1120) . The instructions 1124 may also be referred to as program code or a computer program. The memory 1122 may also store data used by, and results computed by, the processor (s) 1120.
[0157] The network device 1118 may include one or more transceiver (s) 1126 that may include RF transmitter circuitry and / or receiver circuitry that use the antenna (s) 1128 of the network device 1118 to facilitate signaling (e.g., the signaling 1134) to and / or from the network device 1118 with other devices (e.g., the wireless device 1102) according to corresponding RATs.
[0158] The network device 1118 may include one or more antenna (s) 1128 (e.g., one, two, four, or more) . In embodiments having multiple antenna (s) 1128, the network device 1118 may perform MIMO, digital beamforming, analog beamforming, beam steering, etc., as has been described.
[0159] The network device 1118 may include one or more interface (s) 1130. The interface (s) 1130 may be used to provide input to or output from the network device 1118. For example, a network device 1118 that is a base station may include interface (s) 1130 made up of transmitters, receivers, and other circuitry (e.g., other than the transceiver (s) 1126 / antenna (s) 1128 already described) that enables the base station to communicate with other equipment in a core network, and / or that enables the base station to communicate with external networks, computers, databases, and the like for purposes of operations, administration, and maintenance of the base station or other equipment operably connected thereto.
[0160] The network device 1118 may include a PMI decoding module 1132. The PMI decoding module 1132 may be implemented via hardware, software, or combinations thereof. For example, the PMI decoding module 1132 may be implemented as a processor, circuit, and / or instructions 1124 stored in the memory 1122 and executed by the processor (s) 1120. In some examples, the PMI decoding module 1132 may be integrated within the processor (s) 1120 and / or the transceiver (s) 1126. For example, the PMI decoding module 1132 may be implemented by a combination of software components (e.g., executed by a DSP or a general processor) and hardware components (e.g., logic gates and circuitry) within the processor (s) 1120 or the transceiver (s) 1126.
[0161] The PMI decoding module 1132 may be used for various aspects of the present disclosure, for example, aspects of any of FIG. 1, FIG. 2, FIG. 3, FIG. 4, FIG. 5, FIG. 6, FIG. 7, FIG. 9 and FIG. 10. The PMI decoding module 1132 is configured to receive, from a UE, UCI including a first CSI report of CSI reporting where the CSI report includes a first encoded PMI corresponding to the CSI-RS. The PMI decoding module 1132 is further configured to decode, at the base station, a PMI from the encoded PMI by applying at least a first portion of the encoded PMI with a first decoder of the AI / ML model, wherein the first decoder of the AI / ML model uses information corresponding to prior PMIs decoded by the base station to perform decoding for the PMI. Additionally, the PMI decoding module 1132 is configured to select a precoder for use with the UE based on the PMI.
[0162] For one or more embodiments, at least one of the components set forth in one or more of the preceding figures may be configured to perform one or more operations, techniques, processes, and / or methods as set forth herein. For example, a baseband processor as described herein in connection with one or more of the preceding figures may be configured to operate in accordance with one or more of the examples set forth herein. For another example, circuitry associated with a UE, base station, network element, etc. as described above in connection with one or more of the preceding figures may be configured to operate in accordance with one or more of the examples set forth herein.
[0163] Any of the above described embodiments may be combined with any other embodiment (or combination of embodiments) , unless explicitly stated otherwise. The foregoing description of one or more implementations provides illustration and description, but is not intended to be exhaustive or to limit the scope of embodiments to the precise form disclosed. Modifications and variations are possible in light of the above teachings or may be acquired from practice of various embodiments.
[0164] Embodiments and implementations of the systems and methods described herein may include various operations, which may be embodied in machine-executable instructions to be executed by a computer system. A computer system may include one or more general-purpose or special-purpose computers (or other electronic devices) . The computer system may include hardware components that include specific logic for performing the operations or may include a combination of hardware, software, and / or firmware.
[0165] It should be recognized that the systems described herein include descriptions of specific embodiments. These embodiments can be combined into single systems, partially combined into other systems, split into multiple systems or divided or combined in other ways. In addition, it is contemplated that parameters, attributes, aspects, etc. of one embodiment can be used in another embodiment. The parameters, attributes, aspects, etc. are merely described in one or more embodiments for clarity, and it is recognized that the parameters, attributes, aspects, etc. can be combined with or substituted for parameters, attributes, aspects, etc. of another embodiment unless specifically disclaimed herein.
[0166] It is well understood that the use of personally identifiable information should follow privacy policies and practices that are generally recognized as meeting or exceeding industry or governmental requirements for maintaining the privacy of users. In particular, personally identifiable information data should be managed and handled so as to minimize risks of unintentional or unauthorized access or use, and the nature of authorized use should be clearly indicated to users.
[0167] Although the foregoing has been described in some detail for purposes of clarity, it will be apparent that certain changes and modifications may be made without departing from the principles thereof. It should be noted that there are many alternative ways of implementing both the processes and apparatuses described herein. Accordingly, the present embodiments are to be considered illustrative and not restrictive, and the description is not to be limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.
Claims
1.A method of a user equipment (UE) , comprising:receiving, from a base station, a channel state information (CSI) reporting configuration that identifies an artificial intelligence (AI) / machine learning (ML) model and that configures the UE to perform CSI reporting using the AI / ML model;acquiring a precoding matrix indicator (PMI) for a channel between the UE and the base station using a channel state information reference signal (CSI-RS) transmitted by the base station;encoding, at the UE, an encoded PMI by applying at least a first portion of the PMI with a first encoder of the AI / ML model, wherein the first encoder of the AI / ML model uses information corresponding to one or more prior PMIs acquired by the UE to perform encoding for the encoded PMI; andtransmitting, to the base station, uplink control information (UCI) comprising a first CSI report of the CSI reporting, the first CSI report comprising the encoded PMI.2.The method of claim 1, wherein the first encoder of the AI / ML model comprises a recursive neural network (RNN) that is used to encode the encoded PMI and that stores the information corresponding to the one or more prior PMIs as weighting values.3.The method of claim 1, wherein the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting on one of a PUCCH and a PUSCH as part of the CSI reporting.4.The method of claim 1, wherein the CSI report configuration further configures the UE for channel quality indicator (CQI) use in the CSI reporting, and wherein the first CSI report further comprises a CQI acquired by the UE using the CSI-RS.5.The method of claim 1, wherein the CSI report configuration further configures the UE for rank indicator (RI) use in the CSI reporting, and wherein the first CSI report further comprises a RI that is equal to a prior RI reported corresponding to a prior use of the AI / ML model.6.The method of claim 1, further comprising transmitting, to the base station, capability information indicating that the UE can use the AI / ML model.7.The method of claim 6, wherein the capability information is sent to the base station as one of:a UE capability report message; anda radio resource control (RRC) reconfiguration complete message that is sent as part of a handover of the UE to the base station.8.The method of claim 1, further comprising receiving, from the base station, a medium access control control element (MAC CE) that triggers the UE to activate the CSI reporting.9.The method of claim 1, further comprising receiving, from the base station, a medium access control control element (MAC CE) that resets the first encoder to an encoder initial state.10.The method of claim 1, further comprising receiving, from the base station, downlink control information (DCI) that triggers the UE to activate the CSI reporting.11.The method of claim 1, further comprising receiving, from the base station, downlink control information (DCI) that resets the first encoder to an encoder initial state.12.The method of claim 1, wherein the UE further encodes the encoded PMI by applying a second portion of the PMI with a second encoder of the AI / ML model that does not use the information corresponding to the one or more prior PMIs.13.The method of claim 1, further comprising:determining that a UCI container size used by the UE is not large enough to contain both the first CSI report and a second CSI report that does not include layer 1 reference signal receive power (L1-RSRP) information or layer 1 signal to noise and interference (L1-SINR) information; anddropping the second CSI report.14.The method of claim 1, further comprising:receiving, from the base station, a retransmission trigger for the first CSI report; andretransmitting, to the base station, the first CSI report.15.The method of claim 14, wherein:the first CSI report corresponds to an index according to the CSI reporting configuration;the retransmission trigger identifies the CSI reporting configuration and the index corresponding to the first CSI report; andthe UE identifies the first CSI report for retransmission using the index.16.The method of claim 14, wherein the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting as part of the CSI reporting, and wherein UE retransmits the first CSI report separately from the semi-persistent CSI reporting.17.The method of claim 1, further comprising:receiving, from the base station, a retransmission trigger for the first CSI report;determining that a UCI container size used by the UE is not large enough to contain both a retransmission of the first CSI report and a second CSI report of the CSI reporting; anddropping the retransmission of the first CSI report.18.A method of a base station, comprising:sending, to a user equipment (UE) , a channel state information (CSI) reporting configuration that identifies an artificial intelligence (AI) / machine learning (ML) model and that configures the UE to perform CSI reporting using the AI / ML model;transmitting, to the UE, a channel state information reference signal (CSI-RS) ;receiving, from the UE, uplink control information (UCI) comprising a first CSI report of the CSI reporting, the first CSI report comprising a first encoded precoding matrix indicator (PMI) corresponding to the CSI-RS;decoding, at the base station, a PMI from the encoded PMI by applying at least a first portion of the encoded PMI with a first decoder of the AI / ML model, wherein the first decoder of the AI / ML model uses information corresponding to one or more prior PMIs decoded by the base station to perform decoding for the PMI; andselecting a precoder for use with the UE based on the PMI.19.The method of claim 18, wherein the first decoder comprises a recursive neural network (RNN) that is used to decode the encoded PMI and that stores the information corresponding to the one or more prior PMIs as weighting values.20.The method of claim 18, wherein the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting on one of a PUCCH and a PUSCH as part of the CSI reporting.21.The method of claim 18, wherein the CSI report configuration further configures the UE for channel quality indicator (CQI) use in the CSI reporting, and wherein the first CSI report further comprises a CQI acquired by the UE using the CSI-RS.22.The method of claim 18, wherein the CSI report configuration further configures the UE for rank indicator (RI) use in the CSI reporting, and wherein the first CSI report further comprises a RI that is equal to a prior RI reported corresponding to a prior use of the AI / ML model.23.The method of claim 18, further comprising receiving, from the UE, capability information indicating that the UE can use the AI / ML model.24.The method of claim 23, wherein the capability information is received from the UE as one of:a UE capability report message; anda radio resource control (RRC) reconfiguration complete message that is sent as part of a handover of the UE to the base station.25.The method of claim 18, further comprising sending, to the UE, a medium access control control element (MAC CE) that triggers the UE to activate the CSI reporting.26.The method of claim 18, further comprising:determining that the first decoder is out of sync with an encoder of the AI / ML model used by the UE to generate the encoded PMI;sending, to the UE, a medium access control control element (MAC CE) that instructs the UE to reset the encoder to an encoder initial state; andresetting the first decoder to a decoder initial state.27.The method of claim 18, further comprising sending, to the UE, downlink control information (DCI) that triggers the UE to activate the CSI reporting.28.The method of claim 18, further comprising:determining that the first decoder is out of sync with an encoder of the AI / ML model used by the UE to generate the encoded PMI;sending, to the UE, downlink control information (DCI) that instructs the UE to reset the encoder to an encoder initial state; andresetting the first decoder to a decoder initial state.29.The method of claim 18, wherein the base station further decodes the PMI by applying a second portion of the encoded PMI with a second decoder of the AI / ML model that does not use the information corresponding to the one or more prior PMIs.30.The method of claim 18, further comprising:determining to instruct the UE to retransmit the first CSI report;sending, to the UE, a retransmission trigger for the first CSI report; andreceiving, from the UE, a retransmission of the first CSI report.31.The method of claim 30, wherein the base station determines to instruct the UE to retransmit the first CSI report based on a cyclic redundancy check (CRC) of the UCI.32.The method of claim 30, wherein:the first CSI report corresponds to an index according to the CSI reporting configuration; andthe retransmission trigger identifies the CSI reporting configuration and the index corresponding to the first CSI report.33.The method of claim 30, wherein the CSI reporting configuration configures the UE to perform semi-persistent CSI reporting as part of the CSI reporting, and wherein the retransmission of the first CSI report is received separately from the semi-persistent CSI reporting.34.An apparatus comprising means to perform the method of claim 1 to claim 33.35.A computer-readable media comprising instructions to cause an electronic device, upon execution of the instructions by one or more processors of the electronic device, to perform the method of claim 1 to claim 33.36.An apparatus comprising logic, modules, or circuitry to perform the method of claim 1 to claim 33.37.A baseband processor for a user equipment (UE) that is configured to perform one or more elements of claim 1 to claim 17.38.A baseband processor for a base station that is configured to perform one or more elements of claim 18 to claim 33.
Citation Information
Patent Citations
Channel characteristic information transmission method and device, terminal and network side equipment
CN116939650A
Method and apparatus for supporting machine learning or ai ari techniques for CSI feedback in fdd MIMO systems
CN117044125A
Performance monitoring method and device of artificial intelligence AI model, communication equipment, communication system and storage medium
CN117136529A
Method for utilizing channel sparsity and coherence in CSI feedback
US20220417779A1
Method and device for transmitting / receiving wireless signal in wireless communication system
WO2023090762A1