Data transfer configuration using status indication

Abstract

Certain aspects of the present disclosure provide techniques for configuring data transmission. Various aspects relate to determining a data transmission configuration using a machine learning-based algorithm, such as a data transmission configuration for an ultra-reliable low-latency communication (URLLC) application. A method that can be performed by a base station (BS) includes: receiving a feedback report including an indication of a first state corresponding to a plurality of channel condition parameters from a user equipment (UE), and determining one or more actions based at least in part on the first state. The BS can determine one or more actions using a machine learning algorithm using a second state, wherein the second state is based at least in part on the first state.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Application No. 17 / 482,121, filed on September 22, 2021, which claims the benefit of and priority to U.S. Provisional Application No. 63 / 082,436, filed on September 23, 2020, both of which are hereby assigned to the assignee of the present application and are hereby expressly incorporated herein by reference in their entirety, as if fully set forth below and for all applicable purposes. Technical Field

[0003] Aspects of the present disclosure relate to wireless communications and, more particularly, to techniques for configuring data transmissions. Background Art

[0004] Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, broadcast, etc. These wireless communication systems may employ multiple access technologies capable of supporting communication with multiple users by sharing available system resources (e.g., bandwidth, transmit power, etc.). Examples of such multiple access systems include 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) systems, LTE Advanced (LTE-A) systems, Code Division Multiple Access (CDMA) systems, Time Division Multiple Access (TDMA) systems, Frequency Division Multiple Access (FDMA) systems, Orthogonal Frequency Division Multiple Access (OFDMA) systems, Single Carrier Frequency Division Multiple Access (SC-FDMA) systems, and Time Division Synchronous Code Division Multiple Access (TD-SCDMA) systems, to name a few.

[0005] These multiple access technologies have been adopted in various telecommunication standards to provide a common protocol that enables different wireless devices to communicate at a city, country, region, or even global level. New radio (e.g., 5G NR) is an example of an emerging telecommunication standard. NR is a set of enhancements to the LTE mobile standard released by 3GPP. NR is designed to better support mobile broadband Internet access by improving spectral efficiency, reducing costs, improving services, utilizing new spectrum, and better integrating with other open standards using OFDMA with cyclic prefix (CP) on downlink (DL) and uplink (UL). To this end, NR supports beamforming, multiple-input multiple-output (MIMO) antenna technology, and carrier aggregation.

[0006] However, as the demand for mobile broadband access continues to increase, there is a need for further improvements to NR and LTE technologies. Preferably, these improvements should be applicable to other multiple access technologies and the telecommunication standards that employ these technologies. Summary of the invention

[0007] The systems, methods, and devices of the disclosure each have several aspects, no single one of which is solely responsible for its desirable attributes.

[0008] Features of the present disclosure provide advantages including improved data transmission configurations in Ultra-Reliable Low Latency Communication (URLLC) applications by utilizing machine learning based algorithms.

[0009] Certain aspects of the subject matter described in the present disclosure may be implemented in a method for wireless communication by a base station (BS). The method generally includes receiving a feedback report including an indication of a first state corresponding to a plurality of channel condition parameters from a user equipment (UE). The method generally includes determining one or more actions based at least in part on the first state.

[0010] Certain aspects of the subject matter described in the present disclosure may be implemented in a method for wireless communication by a UE. The method generally includes: receiving signaling from a BS for configuring a mapping of a state index to a channel condition parameter. The method generally includes: monitoring a transmission from the BS. The method generally includes: sending a feedback report for the monitored transmission to the BS, the feedback report including an indication of a first state index corresponding to a plurality of channel condition parameters based on the mapping.

[0011] Other aspects provide: an apparatus operable, configured, or otherwise adapted to perform the aforementioned method and those described elsewhere herein; a non-transitory computer-readable medium comprising instructions, which, when executed by one or more processors of the apparatus, causes the apparatus to perform the aforementioned method and those described elsewhere herein; a computer program product on a computer-readable storage medium, comprising code for performing the aforementioned method and the methods described elsewhere herein; and an apparatus comprising units for performing the aforementioned method and those described elsewhere herein. For example, an apparatus may include a processing system, a device having a processing system, or a processing system cooperating on one or more networks.

[0012] To achieve the foregoing and related ends, the one or more aspects include the features fully described below and particularly pointed out in the claims. The following description and the accompanying drawings set forth in detail certain illustrative features of the one or more aspects. However, these features are only indicative of some of the various ways in which the principles of the various aspects can be employed. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order that the above-mentioned features of the present disclosure may be understood in detail, a more specific description briefly summarized above may be obtained by reference to some of the various aspects shown in the accompanying drawings. It should be noted, however, that the drawings only illustrate certain aspects of the present disclosure and that the description may allow for other equally effective aspects.

[0014] Figure 1 is a block diagram conceptually illustrating an example wireless communication network in accordance with certain aspects of the present disclosure.

[0015] Figure 2 is a block diagram conceptually illustrating aspects of an example base station (BS) and user equipment (UE) according to certain aspects of the present disclosure.

[0016] Figure 3 is an example frame format for a particular wireless communication system according to certain aspects of the present disclosure.

[0017] Figure 4 An example hybrid automatic repeat request (HARQ) timeline is shown in accordance with aspects of the present disclosure.

[0018] Figure 5 is a call flow diagram illustrating example signaling for utilizing a machine learning based algorithm to determine a data transmission configuration in accordance with aspects of the present disclosure.

[0019] Figure 6 An example networking environment using a predictive model according to certain aspects of the present disclosure is shown.

[0020] Figure 7 An example reinforcement learning model according to certain aspects of the present disclosure is conceptually illustrated.

[0021] Figure 8 Example nodes in a networking environment are shown in which a predictive model is used to determine a transmission configuration in accordance with certain aspects of the present disclosure.

[0022] Fig. 9 is a flow chart illustrating example operations for wireless communications by a BS in accordance with certain aspects of the present disclosure.

[0023] Fig.10 is a flow chart illustrating example operations for wireless communications by a UE in accordance with certain aspects of the present disclosure.

[0024] Fig.11 A communication device according to aspects of the present disclosure is shown.

[0025] Fig.12 A communication device according to aspects of the present disclosure is shown.

[0026] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. It is contemplated that elements disclosed in one aspect may be beneficially utilized on other aspects without specific recitation. DETAILED DESCRIPTION

[0027] Aspects of the present disclosure provide apparatus, methods, processing systems, and computer-readable media for determining a data transmission configuration. In some examples, a machine learning-based algorithm is used to determine the data transmission configuration. In some examples, the transmission configuration is a data transmission configuration for an ultra-reliable low-latency communication (URLLC) application.

[0028] The state can be associated with multiple channel conditions. A set of channel conditions can be mapped to a state index. The mapping of the state index to the set of channel conditions can be configured at a user equipment (UE). The UE can use the state index to indicate the currently observed set of channel conditions. Based on the state index, the base station (BS) can determine one or more actions associated with the transmission configuration. The BS can use machine learning to determine the transmission configuration and / or the actions to be taken (e.g., configure the transmission configuration). In some examples, the BS can receive an indication of a first state index from the UE in a feedback report, and determine one or more actions based at least in part on the first state index. The BS can use a machine learning algorithm to predict a data transmission configuration that produces a desired reward parameter, or in other words, produces a maximum objective function or a minimum objective function, which helps maintain the desired URLLC reliability requirements.

[0029] The following description provides an example of a data transmission configuration for URLLC applications using a machine learning-based algorithm in a communication system. Without departing from the scope of the present disclosure, the functions and arrangements of the elements discussed may be changed. Various examples may omit, replace, or add various processes or components as appropriate. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted, or combined. In addition, the features described for some examples may be combined in some other examples. For example, any number of aspects set forth herein may be used to implement a device or practice method. In addition, the present disclosure is intended to cover such devices or methods that are practiced using other structures, functions, or structures and functions that are additional to or different from the various aspects of the present disclosure set forth herein. It should be understood that any aspect of the present disclosure disclosed herein may be embodied by one or more elements of the claims.

[0030] In general, any number of wireless networks may be deployed in a given geographic area. Each wireless network may support a specific radio access technology (RAT) and may operate on one or more frequencies. RAT may also be referred to as a radio technology, air interface, etc. Frequency may also be referred to as a carrier, subcarrier, frequency channel, frequency tone, subband, etc. Each frequency may support a single RAT in a given geographic area to avoid interference between wireless networks of different RATs.

[0031] The techniques described herein can be used for various wireless networks and radio technologies. Although various aspects may be described herein using terms typically associated with 3G, 4G, and / or new radio (e.g., 5G NR) wireless technologies, various aspects of the present disclosure may be applied to communication systems based on other generations.

[0032] NR access can support various wireless communication services, such as enhanced mobile broadband (eMBB) targeting wide bandwidth, millimeter wave mmW, massive machine type communication MTC (mMTC) targeting non-backward compatible MTC technology, and / or critical missions targeting URLLC. These services may include latency and reliability requirements. These services may also have different transmission time intervals (TTIs) to meet corresponding quality of service (QoS) requirements. In addition, these services can coexist in the same subframe.

[0033] The electromagnetic spectrum is typically subdivided into various categories, bands, channels, etc. based on frequency / wavelength. In 5G NR, two initial operating bands have been identified with the frequency range names FR1 (410 MHz-7.125 GHz) and FR2 (24.25 GHz-52.6 GHz). Frequencies between FR1 and FR2 are typically referred to as mid-band frequencies. Although a portion of FR1 is greater than 6 GHz, FR1 is often referred to (interchangeably) as the “sub-6 GHz” band in various literature and articles. Similar naming issues sometimes arise with respect to FR2, which is often referred to (interchangeably) as the “millimeter wave” band in literature and articles, although it is different from the extremely high frequency (EHF) band (30 GHz–300 GHz) identified as the “millimeter wave” band by the International Telecommunication Union (ITU).

[0034] In view of the above aspects, unless otherwise specifically stated, it should be understood that the term "sub-6GHz" or similar terms (if used herein) can broadly represent frequencies that can be less than 6GHz, frequencies that can be within FR1, or frequencies that can include mid-band frequencies. In addition, unless otherwise specifically stated, it should be understood that the term "millimeter wave" and the like (if used herein) can broadly represent frequencies that can include mid-band frequencies, frequencies that can be within FR2, or frequencies that can be within the EHF band.

[0035] NR supports beamforming and can dynamically configure the beam direction. MIMO transmission with precoding can also be supported. MIMO configuration in DL can support up to 8 transmit antennas with multi-layer DL transmission of up to 8 streams and up to 2 streams per UE. Multi-layer transmission with up to 2 streams per UE can be supported. Aggregation of multiple cells with up to 8 serving cells can be supported.

[0036] Figure 1 An example wireless communication network 100 is shown in which aspects of the present disclosure may be performed. For example, the wireless communication network 100 may be a NR system (e.g., a 5G NR network). Figure 1 As shown, the wireless communication network 100 may communicate with a core network 132. The core network 132 may communicate with one or more BSs 110a-z (each also individually referred to herein as BS 110 or collectively referred to herein as BS 110) and / or UEs 120a-y (each also individually referred to herein as UE 120 or collectively referred to herein as UE 120) in the wireless communication network 100 via one or more interfaces.

[0037] According to certain aspects, BS 110 and UE 120 may be configured for data transmission in URLLC applications. Figure 1 As shown, BS 110a includes a data transmission configuration manager 112. The data transmission configuration manager 112 may be configured to receive a first state from UE 120a in a feedback report, the first state corresponding to a plurality of channel conditions. According to aspects of the present disclosure, the data transmission configuration manager 112 may be configured to determine one or more actions based at least in part on the first state. Figure 1 As shown, UE 120a includes a data transmission configuration manager 122. According to aspects of the present disclosure, data transmission configuration manager 122 may be configured to monitor transmissions and send a feedback report for the monitored transmissions to BS 110a, the feedback report including an indication of the first state.

[0038] BS 110 may provide communication coverage for a particular geographic area (sometimes referred to as a "cell"), which may be stationary or may move depending on the location of mobile BS 110. In some examples, BS 110 may be interconnected to each other and / or to one or more other BSs or network nodes (not shown) in wireless communication network 100 through various types of backhaul interfaces (e.g., direct physical connections, wireless connections, virtual networks, etc.) using any suitable transport network. Figure 1In the example shown, BSs 110a, 110b, and 110c may be macro BSs for macro cells 102a, 102b, and 102c, respectively. BS 110x may be a pico BS for pico cell 102x. BSs 110y and 110z may be femto BSs for femto cells 102y and 102z, respectively. A BS may support one or more cells.

[0039] UEs 120 (e.g., 120x, 120y, etc.) may be dispersed throughout the wireless communication network 100, and each UE 120 may be stationary or mobile. The wireless communication network 100 may also include a relay station (e.g., relay station 110r) (also referred to as a relay, etc.) that receives transmissions of data and / or other information from an upstream station (e.g., BS 110a or UE 120r) and sends transmissions of the data and / or other information to a downstream station (e.g., UE 120 or BS 110), or relays transmissions between UEs 120 to facilitate communication between devices.

[0040] The network controller 130 may communicate with a set of BSs 110 and provide coordination and control (e.g., via a backhaul) for these BSs 110. In various aspects, the network controller 130 may communicate with a core network 132 (e.g., a 5G core network (5GC)), which provides various network functions such as access and mobility management, session management, user plane functions, policy control functions, authentication server functions, unified data management, application functions, network exposure functions, network repository functions, network slice selection functions, and the like.

[0041] Figure 2 1 shows a BS 110a and a UE 120a (eg, Figure 1 Example components of a wireless communication network 100).

[0042] At BS 110a, transmit processor 220 may receive data from data source 212 and control information from controller / processor 240. The control information may be for a physical broadcast channel (PBCH), a physical control format indicator channel (PCFICH), a physical hybrid ARQ indicator channel (PHICH), a physical downlink control channel (PDCCH), a group common PDCCH (GC PDCCH), etc. The data may be for a physical downlink shared channel (PDSCH), etc. A medium access control (MAC)-control element (MAC-CE) is a MAC layer communication structure that may be used for control command exchange between wireless nodes. The MAC-CE may be carried in a shared channel, such as a physical downlink shared channel (PDSCH), a physical uplink shared channel (PUSCH), or a physical sidelink shared channel (PSSCH).

[0043] The processor 220 may process (e.g., encode and symbol map) data and control information to obtain data symbols and control symbols, respectively. The transmit processor 220 may also generate reference symbols, for example, for a primary synchronization signal (PSS), a secondary synchronization signal (SSS), a PBCH demodulation reference signal (DMRS), and a channel state information reference signal (CSI-RS). The transmit (TX) multiple-input multiple-output (MIMO) processor 230 may perform spatial processing (e.g., precoding) on ​​data symbols, control symbols, and / or reference symbols (if applicable), and may provide an output symbol stream to a modulator (MOD) in a transceiver 232a-232t. Each modulator in the transceiver 232a-232t may process a respective output symbol stream (e.g., for OFDM, etc.) to obtain an output sample stream. Each modulator may further process (e.g., convert to analog, amplify, filter, and up-convert) the output sample stream to obtain a downlink signal. Downlink signals from modulators in transceivers 232a-232t may be transmitted via antennas 234a-234t, respectively.

[0044] At UE 120a, antennas 252a-252r may receive downlink signals from BS110a and may provide received signals to demodulators (DEMODs) in transceivers 254a-254r, respectively. Each demodulator in transceivers 254a-254r may adjust (e.g., filter, amplify, downconvert, and digitize) respective received signals to obtain input samples. Each demodulator may further process the input samples (e.g., for OFDM, etc.) to obtain received symbols. MIMO detector 256 may obtain received symbols from all demodulators in transceivers 254a-254r, perform MIMO detection on the received symbols (if applicable), and provide detected symbols. Receive processor 258 may process (e.g., demodulate, deinterleave, and decode) detected symbols, provide decoded data to UE 120a to data sink 260, and provide decoded control information to controller / processor 280.

[0045] On the uplink, at the UE 120a, a transmit processor 264 may receive and process data from a data source 262 (e.g., for a physical uplink shared channel (PUSCH)) and control information from a controller / processor 280 (e.g., for a physical uplink control channel (PUCCH)). The transmit processor 264 may also generate reference symbols for reference signals (e.g., for a sounding reference signal (SRS)). The symbols from the transmit processor 264 may be precoded by a TX MIMO processor 266 (if applicable), further processed by a modulator in a transceiver 254a-254r (e.g., for SC-FDM, etc.), and sent to the BS 110a. At BS 110a, the uplink signal from UE 120a may be received by antenna 234, processed by demodulators in transceivers 232a-232t, detected by MIMO detector 236 (if applicable), and further processed by receive processor 238 to obtain decoded data and control information sent by UE 120a. Receive processor 238 may provide the decoded data to a data sink 239 and the decoded control information to controller / processor 240.

[0046] Memories 242 and 282 may store data and program codes for BS 110a and UE 120a, respectively.A scheduler 244 may schedule UEs for data transmission on the downlink and / or uplink.

[0047] The antenna 252, processors 266, 258, 264 and / or controller / processor 280 of the UE 120a and / or the antenna 234, processors 220, 230, 238 and / or controller / processor 240 of the BS 110a may be used to perform the various techniques and methods described herein. Figure 2 As shown in FIG. 1 , the controller / processor 240 of BS 110 a has a data transmission configuration manager 241 (e.g., which may be Figure 1 112), and the controller / processor 280 of the UE 120a has a data transmission configuration manager 281 (e.g., which may be Figure 1 Although illustrated at the controller / processor level, other components of the UE 120a and BS 110a may be used to perform the operations described herein.

[0048] NR can utilize orthogonal frequency division multiplexing (OFDM) with cyclic prefix (CP) on uplink and downlink. NR can support half-duplex operation using time division duplex (TDD). OFDM and single carrier frequency division multiplexing (SC-FDM) divide the system bandwidth into multiple orthogonal subcarriers, which are also commonly referred to as frequency modulation, frequency band, etc. Each subcarrier can be modulated with data. Modulation symbols can be sent in the frequency domain using OFDM, and modulation symbols can be sent in the time domain using SC-FDM. The spacing between adjacent subcarriers can be fixed, and the total number of subcarriers can depend on the system bandwidth. The minimum resource allocation, called a resource block (RB), can be 12 consecutive subcarriers. The system bandwidth can also be divided into subbands. For example, a subband can cover multiple RBs. NR can support a basic subcarrier spacing (SCS) of 15kHz, and other SCSs (e.g., 30kHz, 60kHz, 120kHz, 240kHz, etc.) can be defined for the basic SCS.

[0049] Figure 3 300 is a schematic diagram showing an example of a frame format for NR. The transmission timeline of each of the downlink and uplink can be divided into units of radio frames. Each radio frame can have a predetermined duration (e.g., 10ms) and can be divided into 10 subframes of 1ms each with an index of 0 to 9. Each subframe can include a variable number of slots (e.g., 1, 2, 4, 8, 16, ... slots) depending on the SCS. Each slot can include a variable number of symbol periods (e.g., 7, 12 or 14 symbols) depending on the SCS. An index can be assigned to the symbol period in each slot. The subslot structure can refer to a transmission time interval having a duration less than a slot (e.g., 2, 3 or 4 symbols). Each symbol in a slot can be configured for a link direction (e.g., DL, UL or flexible) of data transmission, and the link direction of each subframe can be dynamically switched. The link direction can be based on the slot format. Each slot can include DL / UL data and DL / UL control information.

[0050] As described above, NR access can support various wireless communication services, including URLLC. URLLC can be targeted at services that require both high reliability and low latency, such as traffic control (e.g., control of autonomous vehicles) or remote control (e.g., drones or remote surgery).

[0051] Reliability can be defined as the ability to deliver a certain amount of traffic with a high probability of success within a predetermined duration. To achieve ultra-reliable communication, the URLLC reliability requirement can be 1e -5 Or a maximum block error rate (BLER) of 0.001%. In an example, in a URLLC application, when there are two transmissions and the first TX BLER is 10%, the second TX BLER must be 1e -4 To meet 1e -5 To maintain the desired BLER condition in both transmissions, the UE may feed back to the network a channel state information (CSI) report and a hybrid automatic repeat request (HARQ) acknowledgement (ACK) or negative acknowledgement (NACK) report of the previous transmission of the BS.

[0052] The cellular system includes a mechanism that allows the UE to feed back information about channel conditions to the network to assist the scheduler at the BS. Such feedback may include information about observed BLER, observed channel quality indicator (CQI), observed precoding matrix indicator (PMI), observed best redundancy version (RV), etc. Such information may be sent synchronously with the HARQ-ACK feedback.

[0053] HARQ is an integral part of data communications to ensure reliability and performance. Figure 4 An example HARQ timeline 400 is shown. Figure 4As shown, the feedback of HARQ ACK / NACK information may include separate ACK / NACK information or also have channel status information (e.g., CSI report). The feedback process of HARQ ACK / NACK begins with the BS sending a downlink grant 402 to the UE. The DL grant 402 may be carried in downlink control information (DCI) on the PDCCH. The DL grant 402 may schedule data transmission on the PDSCH 404. The PDSCH 404 may be scheduled at the first DL time slot after the delay period K0 between the DL grant 402 and the data transmission. The UE may provide HARQ-ACK feedback for the PDSCH 404 in the second DL time slot after another delay period K1 between the PDSCH 404 and the HARQ-ACK feedback in the HARQ-ACK feedback report. The UE may send the HARQ-ACK feedback report in uplink control information (UCI) on the PUCCH (or in some cases, multiplexed on the PUSCH). The HARQ-ACK feedback report may be a regular HARQ-ACK report 406 that includes only ACK / NACK feedback and does not include CSI. The HARQ-ACK feedback report may be a soft / tuboo HARQ-ACK feedback report 408 that includes ACK / NACK feedback and CSI reporting. The UE may determine HARQ ACK feedback (e.g., ACK or NACK) based on monitoring the scheduled PDSCH 404. The UE determines whether the PDSCH 404 was successfully decoded. The CSI feedback may be generated based on measurements of the PDSCH, or in some cases, based on measurements of the CSI-RS (e.g., which may be scheduled / activated by the DCI).

[0054] Example Data Transfer Configuration Using Status Indication

[0055] Aspects of the present disclosure provide enhancements in determining data transmission configuration using status indications. In some aspects, the status indications can be used in machine learning-based algorithms to determine data transmission configurations. The data transmission configuration can be determined to maintain the reliability and latency targets of ultra-reliable low-latency communications (URLLC) applications.

[0056] Figure 5 is a call flow diagram illustrating example signaling 500 for determining a data transmission configuration using a status indication in accordance with aspects of the present disclosure. Figure 5As shown in the example of , at 502, a base station (e.g., BS110a) configures a mapping of state indexes to channel state parameters for a user equipment (e.g., UE 120a). The configuration may also indicate the number M of states to be used on the UE side and / or the contents of the M states (e.g., the channel state parameters associated with each state). At 504, a quantization level for the channel state parameters may be configured for the UE 120a. The BS110a may also indicate the quantization level for the channel state parameters to the UE 120a, or the quantization level may be specified in the wireless standard or determined by the UE 120a.

[0057] Channel condition parameters may include block error rate (BLER), log-likelihood ratio (LLR) quality, interference covariance matrix eigenvalues, interference power, the number of code blocks (CBs) that are negatively acknowledged (NACKed), the number of CB groups (CBGs) that are NACKed, a channel quality indicator (CQI), spectral efficiency (SPEF), signal-to-noise ratio (SNR), parity status, observed best redundancy version (RV), current analog precoder used at the UE, current digital precoder used at the UE, best analog precoder observed at the UE, best digital precoder observed at the UE, observed best demodulation reference signal (DMRS) configuration, current interference pattern in frequency, current interference pattern in time, current interference covariance matrix rank, current interference power, and / or optimal resource allocation at the UE.

[0058] The BS may configure the UE via Radio Resource Control (RRC) signaling and / or Medium Access Control (MAC) Control Element (CE) signaling.

[0059] At 506, UE 120a monitors transmission 507 from BS 110a, which may be successfully or unsuccessfully received and decoded by UE 120a. For example, UE 120a may monitor physical downlink shared channel (PDSCH) transmissions (e.g., data transmissions such as PDSCH 404) scheduled by downlink grants (e.g., such as DL grant 402) in downlink control information (DCI) and / or channel state information reference signals (CSI-RS) that may be activated by DCI. UE 120a may measure channel conditions using reception of PDSCH and / or CSI reference signals (CSI-RS). UE 120a may determine a state. For example, UE 120a may determine a state corresponding to a channel condition based on a configured mapping of channel conditions to a state index. In some examples, a soft HARQ-ACK report is mapped to a state index.

[0060] At 508, the UE 120a sends a feedback report for the monitored transmission as part of a hybrid automatic repeat request (HARQ) acknowledgement (ACK) report. The feedback report includes a first state index (associated with the first state), the first state index corresponding to a plurality of channel condition parameters (e.g., based on the monitored transmission at 506) based on a mapping.

[0061] At 510, after BS 110a receives the feedback report including an indication of the first state index, BS 110a may determine a second state (e.g., a "super state") by adding the first state associated with the first state index sent by UE 120a to a state maintained at BS 110a (e.g., a third state). For example, the state maintained by the BS may include a current set of channel condition parameters, such as a current resource allocation, a digital precoder at the BS, an analog precoder at the BS, a current modulation and coding scheme (MCS), a current data rank, a current action, a current demodulation reference signal (DMRS) configuration, and / or a current transmit power. Thus, when BS 110a receives the first state from UE 120a, a plurality of channel condition parameters corresponding to the first state may be added to the third state already maintained at BS 110a to determine a second state (e.g., a super state).

[0062] At 512, BS110a determines one or more actions based on the super state (e.g., the second state determined at 510). One or more actions may be associated with a data transmission configuration. According to certain aspects, BS110a uses a machine learning algorithm to determine one or more actions. For example, BS110a may input the determined super state (e.g., the second state) into the machine learning algorithm. The machine learning algorithm may be modeled as a Markov decision process (MDP). In some examples, the machine learning algorithm may be a reinforcement learning algorithm, a deep learning algorithm, a supervised learning algorithm, an unsupervised learning algorithm, a Q learning algorithm, a value reinforcement algorithm, a polarity reinforcement algorithm, or a combination thereof. BS110a may solve the machine learning algorithm to predict the best action from a predefined action set. The best action may include an action for configuring a data transmission configuration, an action for the BS and / or an action for the UE. For example, the best action may include selecting the following for subsequent transmission: MCS, channel rank, transmit precoder at the BS, first analog precoder at the UE, second analog precoder at the BS, first digital precoder at the UE, second digital precoder at the BS, RV, DMRS configuration, resource allocation, or a combination of the above.

[0063] For each state index, there is some action (e.g., an optimal or best action) that, if performed, will result in a reward parameter, or in other words, a maximum objective function or a minimum objective function. The reward parameter may identify signal throughput, spectral efficiency, minimum BLER, minimum delay, minimum number of retransmissions, target BLER associated with URLLC, target latency associated with URLLC, optimal analog and digital precoders on the UE side, or a combination thereof. The action may be an action for configuring a transmission configuration that is predicted to achieve a reward.

[0064] At 514, BS110a signals one or more actions to UE 120a. These actions may be: one or more actions already taken by BS110a, one or more decided actions to be taken by BS110a, one or more actions to be taken by UE 120a, or a subset of these actions. BS110a may signal these actions via DCI. BS110a may signal one or more actions separately, or signal one or more actions as bundled information. In order to signal one or more actions as bundled information, BS110a may signal the index representing these actions to UE 120a. In some examples, BS110a configures an action set and a mapping to an index for UE 120a via RRC or MAC-CE signaling, and then signals the index corresponding to the action set taken and / or to be taken to UE 120a in the DCI.

[0065] In some examples, such as Figure 5 As shown, one or more actions determined using a machine learning-based algorithm can configure a predicted improved data transmission configuration to achieve a target BLER associated with URLLC. The improved data transmission configuration can be associated with a reward parameter that helps maintain the desired BLER and / or latency target for URLLC.

[0066] Although not in Figure 5 , but utilizing a machine learning based algorithm may involve training a model, such as a prediction model. The model may be used to predict the best action based on the current state. For example, the best action may correspond to an action predicted to be most likely to achieve the desired reward. The model may use the current super state maintained by BS 110a as an input to the model. The model may be used to perform the above predictions to determine the data transmission configuration. The model may be trained based on training data (e.g., training information), which may include channel condition parameters, states, actions, and associated rewards.

[0067] Figure 66 shows an example networking environment 600 in which a data transmission configuration manager 622 of a BS 110 uses a prediction model 624 to determine a data transmission configuration according to certain aspects of the present disclosure. Figure 6 As shown, the networked environment 600 includes the BS 110, the training system 630, and the training repository 615 that are communicatively connected via one (or more) networks 605. The one (or more) networks 605 may include a wireless network such as the wireless communication network 100. Although the training system 630, the BS 110a, and the training repository 615 are in Figure 6 Although shown as separate components in FIG. 6 , the training system 630 , BS 110 a , and / or the training repository 615 may be implemented as one or more stand-alone systems or in a distributed environment on any number of computing systems.

[0068] The training system 630 generally includes a prediction model training manager 632 that uses training data to generate a prediction model 624 for predicting a data transmission configuration that achieves a desired reward based on a current state. The prediction model 624 can be trained based at least in part on information in the training repository 615.

[0069] The training repository 615 may include training data obtained before and / or after the deployment of the BS 110a. Before the deployment of the BS 110a, the BS 110a may be trained in a simulated communication environment (e.g., in a field test, a drive test). This information may be stored in the training repository 615. After deployment, the training repository 615 may be updated to include feedback associated with the data transmission configuration used by the BS 110a. The training repository 615 may also be updated with information from other BSs and / or other UEs, for example, based on the experience learned by those BSs and UEs.

[0070] The prediction model training manager 632 may use the information in the training repository 615 to determine (e.g., generate) a prediction model 624 for determining a data transmission configuration. The prediction model training manager 632 may use a variety of different types of machine learning algorithms to form the prediction model 624. The training system 630 may be located on the BS 110a or on a different entity for determining the prediction model 624. If located on a different entity, the prediction model 624 is provided to the BS 110a. The training repository 615 may be a storage device, such as a memory. The training repository 615 may be located on the BS 110a, on the training system 630, in a cloud storage device, or in another entity in the network 605. The training repository 615 may receive training information from the BS 110a, an entity in the network 605 (e.g., a UE in the network 605), a cloud, or other sources.

[0071] Machine learning can use any suitable machine learning algorithm. In some non-limiting examples, the machine learning algorithm is a reinforcement learning algorithm, a value reinforcement algorithm, a supervised learning algorithm, an unsupervised learning algorithm, a deep learning algorithm, an artificial neural network algorithm, a Q learning algorithm, a polarity reinforcement algorithm, or other types of machine learning algorithms.

[0072] In some examples, a deep convolutional network (DCN) is used to perform machine learning (e.g., used by training system 630). A DCN is a network of convolutional networks, configured with additional pooling and normalization layers. DCN has achieved state-of-the-art performance on many tasks. A DCN can be trained using supervised learning, in which the input and output targets are known for many samples and are used to modify the weights of the network by using a gradient descent method. A DCN can be a feedforward network. In addition, as described above, the connection from a certain neuron in the first layer of a DCN to a neuron group in the next higher layer is shared between the neurons in the first layer. The feedforward and shared connections of a DCN can be used for fast processing. The computational burden of a DCN can be, for example, much smaller than the computational burden of a neural network of similar size including recursive or feedback connections.

[0073] In some examples, neural networks are used to perform machine learning (e.g., used by training system 630). Neural networks can be designed to have various connection patterns. In a feedforward network, information is passed from a lower layer to a higher layer, where each neuron in a given layer communicates with several neurons in a higher layer. Hierarchical representations can be constructed in successive layers of a feedforward network. Neural networks can also have recursive or feedback (also referred to as top-down) connections. In a recursive connection, the output from a neuron in a given layer can be communicated to another neuron in the same layer. Recursive architectures can help identify patterns that span more than one input data block that is sequentially delivered to a neural network. The connection from a neuron in a given layer to a neuron in a lower layer is called a feedback (or top-down) connection. When the recognition of high-level concepts can help distinguish specific low-level features of the input, a network with many feedback connections may be helpful.

[0074] An artificial neural network, which may be composed of an interconnected group of artificial neurons (e.g., a neuron model), is a computing device or represents a method performed by a computing device. These neural networks can be used in various applications and / or devices, such as Internet Protocol (IP) cameras, Internet of Things (IoT) devices, autonomous vehicles, and / or service robots. Each node in an artificial neural network can simulate a biological neuron by acquiring input data and performing simple operations on the data. The results of the simple operations performed on the input data are selectively passed to other neurons. Weight values ​​are associated with each vector and node in the network, and these values ​​constrain how the input data is related to the output data. For example, the input data of each node can be multiplied by the corresponding weight value, and the product can be summed. The sum of the products can be adjusted by an optional bias, and an excitation function can be applied to the result, thereby generating an output signal or "output excitation" of the BS. The weight values ​​can initially be determined by an iterative flow of training data through the network (e.g., during a training phase in which the network learns how to identify a specific category by its typical input data characteristics, the weight values ​​are established).

[0075] Machine learning (e.g., used by the training system 630) can be implemented using different types of artificial neural networks, such as recursive neural networks (RNNs), multilayer perceptron (MLP) neural networks, convolutional neural networks (CNNs), and the like. The working principle of RNNs is to save the output of a layer and feed the output back to the input to help predict the result of the layer. In an MLP neural network, data can be fed into an input layer, and one or more hidden layers provide an abstract level of the data. The output layer can then be predicted based on the abstract data. MLPs may be particularly suitable for classification prediction problems in which a category or label is assigned to the input. A convolutional neural network (CNN) is a feedforward artificial neural network. A convolutional neural network may include a collection of artificial neurons, each of which has a receptive field (e.g., a spatially localized region of the input space) and tiles the input space together. Convolutional neural networks have many applications. In particular, CNNs have been widely used in the fields of pattern recognition and classification. In a layered neural network architecture, the output of the first layer of artificial neurons becomes the input of the second layer of artificial neurons, the output of the second layer of artificial neurons becomes the input of the third layer of artificial neurons, and so on. Convolutional neural networks can be trained to recognize the hierarchy of features. The computation in a convolutional neural network architecture can be distributed across a cluster of processing nodes, which can be configured in one or more computation chains. These multi-layer architectures can be trained one layer at a time and can be fine-tuned using backpropagation.

[0076] In some examples, when a machine learning algorithm is used, the training system 630 generates vectors from the information in the training repository 615. In some examples, the training repository 615 stores vectors. In some examples, the vectors map one or more features to labels. For example, the features may correspond to various one or more actions for data transmission configurations and / or various states. The labels may correspond to various predicted results for those actions and states. The prediction model training manager 632 may use the vectors to train the prediction model 624 for BS110a. As described above, the vectors may be associated with weights in the machine learning algorithm.

[0077] Figure 7 An example reinforcement learning model is conceptually shown. Reinforcement learning can be a semi-supervised learning model in machine learning. Reinforcement learning allows an agent 704 (e.g., BS110a) to take actions (e.g., data transmission configuration) based on states (e.g., super states including states received from UEs) observed by an interpreter 702 (e.g., such as UE 120a), and interact with an environment 706 (e.g., wireless communication network 100) to maximize a total reward (e.g., signal throughput, spectral efficiency, minimum BLER, minimum delay, minimum number of retransmissions, target BLER associated with URLLC, target delay associated with URLLC, optimal analog and digital precoders on the UE side, or a combination thereof), which can be observed by the interpreter 702 and fed back to the agent 704 as an enhancement.

[0078] In some examples, machine learning is modeled as a Markov decision process (MDP). An MDP is a discrete, time-stochastic control process. MDPs provide a mathematical framework for modeling decisions where the outcome may be partially random and partially under the control of the decision maker. In an MDP, at each time step, the process is in one of a set of S finite states, and the decision maker can choose any action available in that state from a finite set of actions A. The process responds at the next time step by randomly moving to a new state and giving the decision maker a corresponding reward, where R α (s, s′) is the immediate reward (or expected immediate reward) after transitioning from state s to state s′. The probability that the process moves to its new state is affected by the action chosen, for example, according to the state transition function. The state transition can be represented by P α (s,s′)=Pr(s t+1 =s′|s t =s,α t =α) is given.

[0079] The MDP seeks to find a policy for making decisions: a function of π that specifies the action π(s) that the decision maker will choose when in state s. The goal is to choose a policy π that maximizes the reward. For example, a policy that maximizes a cumulative function of the reward, such as the discounted sum. An example function is shown below:

[0080] in

[0081] α t =π(s t ), the action given by the policy, and γ is a discount factor and satisfies 0≤γ≤1.

[0082] The solution to an MDP is a policy that describes the best action for each state in the MDP, e.g., the policy that maximizes the expected discounted reward.

[0083] In some examples, a partially observable MDP (POMDP) ​​is used. POMDPs can be used when the state may not be known when an action is taken, and therefore the probability and / or reward may be unknown. For POMDPs, reinforcement learning can be used. The following function can be defined:

[0084] Q(s,a)=∑ s′ P α (s,s′)(R α (s,s′)+γV(s′)),

[0085] Where V(s′) is the discounted cumulative reward.

[0086] Experience during learning can be based on (s, a) pairs and results s'. For example, if the node was previously in state s, and made base station analog / digital precoder, MCS, RV index, DMRS configuration, resource allocation selection (and other actions) a, and achieved throughput s'. In this example, the node can directly update the array Q based on the learned experience. This can be called Q-learning. In some examples, the learning algorithm can be continuous.

[0087] Return to reference Figure 6 The example networking environment 600 and Figure 7In some examples, the prediction model training manager 632 or the agent 704 may use reinforcement learning for the prediction model (e.g., prediction model 624) to determine a policy (e.g., to solve the MDP solution). The BS 110a or the agent 704 may take actions, such as data transmission configurations, based on the policy given by the prediction model (e.g., prediction model 624) at a given time for the current state in the environment (e.g., environment 706) (e.g., observed by the UE 120a or the interpreter 702). The reinforcement learning algorithm and the prediction model may be updated / adapted based on the learned experience (e.g., which may be stored in the training repository 615).

[0088] The framework of reinforcement learning provides tools to optimally solve POMDPs. Learning changes the weights of a multi-level perceptron (e.g., a neural network) that determines the next action to take. The algorithms in deep ML are encoded in the neural network weights. Therefore, changing the weights changes the algorithm.

[0089] In some examples, the machine learning-based data transmission configuration uses a deep learning algorithm. The deep learning algorithm can be a deep Q network (DQN) algorithm implemented by a neural network.

[0090] In some examples, machine learning algorithms are modeled as POMDPs with reinforcement learning. POMDPs can be used when the state may not be known when an action is taken, and therefore the probability and / or reward may be unknown. For POMDPs, reinforcement learning can be used. The Q array can be defined as:

[0091] Q i+1 (s,a)=E{r+γmax Q i (s′,a′)|s,a}.

[0092] According to certain aspects, the machine learning-based data transmission configuration allows for continuous unlimited learning. In some examples, the learning can be enhanced with federated learning. For example, while some machine learning methods use centralized training data on a single machine or in a data center; with federated learning, the learning can be a collaboration involving multiple devices to form a predictive model. With federated learning, the training of the model can be completed on the device using collaborative learning from multiple devices. For example, return to reference Figure 6-7 , BS 110a and agent 704 may receive training information and / or updated training models from a variety of different devices.

[0093] In an illustrative example, a machine learning data transmission configuration algorithm for multiple different BSs can be trained in multiple different operating scenarios, for example, using deep reinforcement learning. The outputs from the training of different BSs can be combined to train a beam management algorithm for the BS. As described above, the state can be a channel condition parameter observed by the UE; the action can be a data transmission configuration; and the reward can be a URLLC reliability and / or latency target.

[0094] like Figure 8 As shown, BS 820 (e.g., such as Figure 6 The BS 110a in the illustrated networking environment 600 may include a data transmission configuration manager 822. The data transmission configuration manager 822 may receive a feedback report (824) including an indication of a first state (e.g., a state index value of the first state) corresponding to a plurality of channel condition parameters from a UE (e.g., UE 120a). The data transmission configuration manager 822 may add the received first state to a third state maintained at the BS 820 to determine a second state (826). The data transmission configuration manager 822 may include a prediction model 828 (e.g., such as the prediction model 624). The data transmission configuration manager uses the prediction model 828 to determine one or more actions based at least in part on the first state (829). The BS 820 may signal to the UE at least one or more actions to be taken by the UE (830).

[0095] Fig. 9 900 is a flow chart illustrating example operations 900 for wireless communication according to certain aspects of the present disclosure. Operations 900 may be performed, for example, by a BS (e.g., such as BS 110a in wireless communication network 100). Operations 900 may be implemented as a processor (e.g., Figure 2 In addition, in operation 900, the BS may transmit and receive signals, for example, by one or more antennas (e.g., Figure 2 In certain aspects, the transmission and / or reception of signals by the BS may be implemented via a bus interface of one or more processors (eg, controller / processor 240) that obtain and / or output signals.

[0096] Operation 900 may begin at 902, receiving a feedback report from a UE including an indication of a first state corresponding to a plurality of channel condition parameters. The feedback report may include a HARQ ACK report, a CSI report, or both for a previous transmission by the BS. In some examples, the plurality of channel condition parameters include one or more of: BLER, number of negatively acknowledged CBs, number of NACKed CBGs, CQI, SPEF, SNR, LLR quality, parity status, observed best RV, current analog precoder used at the UE, current digital precoder used at the UE, best analog precoder observed at the UE, best digital precoder observed at the UE, observed best DMRS configuration, current interference pattern in frequency and / or time, current interference covariance matrix rank, current interference power, current interference covariance matrix eigenvalues, best resource allocation at the UE, or a combination thereof.

[0097] At 904, the BS determines one or more actions based at least in part on the first state. In some examples, determining one or more actions based at least in part on the first state includes: determining one or more actions using a machine learning algorithm using a second state, wherein the second state is based at least in part on the first state. Determining one or more actions using a machine learning algorithm may include determining one or more of the following for subsequent transmission: MCS, channel rank, transmit precoder at the BS, first analog precoder at the UE, second analog precoder at the BS, first digital precoder at the UE, second digital precoder at the BS, RV, DMRS configuration, resource allocation, or a combination thereof. In some examples, machine learning may use states associated with multiple channel conditions, reward parameters, and action parameters associated with one or more parameters associated with subsequent transmission, the reward parameters being associated with the following: signal throughput, spectral efficiency, minimum BLER, minimum delay, minimum number of retransmissions, target BLER associated with URLLC, target delay associated with URLLC, or a combination thereof. In some examples, one or more actions may include one or more actions to be taken by the BS, one or more actions to be taken by the UE, or a combination thereof.

[0098] In some examples, the machine learning algorithm includes a reinforcement learning algorithm, a deep learning algorithm, a supervised learning algorithm, an unsupervised learning algorithm, a Q learning algorithm, a value reinforcement algorithm, a polarity reinforcement algorithm, or a combination thereof. The machine learning algorithm can be modeled as a Markov decision process (MDP).

[0099] In some examples, the machine learning algorithm is implemented by an artificial neural network. The artificial neural network may include a deep Q network (DQN) that includes one or more deep neural networks (DNNs) and determining one or more actions using machine learning may include: passing one or more second states and one or more action parameters through one or more DNNs, outputting a value for each action parameter, and selecting an action associated with an optimal output value.

[0100] Operation 900 may also include: the BS configuring a mapping of a state index to a channel condition parameter for the UE. The indication of the first state is one of the state indexes. In some examples, the configuration is via RRC signaling or MAC-CE signaling.

[0101] Operation 900 may also include: the BS configuring the UE with the number of states to be maintained at the UE, channel condition parameters associated with the states, or both. In some examples, the configuration is via RRC signaling or MAC-CE signaling.

[0102] Operation 900 may also include: the BS configuring, for the UE, a quantization level for one or more channel condition parameters among the plurality of channel condition parameters. In some examples, the configuration is via RRC signaling or MAC-CE signaling.

[0103] Operation 900 may also include: the BS determines a second state by adding the first state to a third state maintained at the BS. In some examples, the third state corresponds to one or more of: a current resource allocation, a digital precoder at the BS, an analog precoder at the BS, a current MCS, a current data rank, a current action, a current DMRS configuration, a current transmit power, or a combination thereof.

[0104] Operation 900 may also include the BS at least signaling to the UE one or more actions to be taken by the UE. The signaling may include signaling each of the one or more actions separately or signaling the one or more actions together via an index mapped to the multiple actions.

[0105] Operation 900 may further include: the BS transmitting RRC signaling or MAC-CE signaling indicating the mapping, and the signaling includes DCI including the index.

[0106] Fig.10 1 is a flow chart illustrating example operations 1000 for wireless communication according to certain aspects of the present disclosure. Operations 1000 may be performed, for example, by a UE (e.g., such as UE 120a in wireless communication network 100). Operations 1000 may be complementary to operations 900 performed by a BS. Operations 1000 may be implemented as a process on one or more processors (e.g., Figure 2 In addition, the UE may transmit and receive signals in operation 1000 by, for example, one or more antennas (e.g., Figure 2 In certain aspects, the UE's transmission and / or reception of signals may be implemented via a bus interface of one or more processors (eg, controller / processor 280) that obtain and / or output signals.

[0107] Operations 1000 may begin when a UE receives signaling from a BS for configuring a mapping of a state index to a channel condition parameter at 1002. In some examples, the configuration is via RRC signaling or MAC-CE signaling.

[0108] At 1004, the UE monitors transmissions from the BS.

[0109] At 1006, the UE sends a feedback report for the monitored transmission to the BS, the feedback report including an indication of a first state index corresponding to a plurality of channel state parameters based on the mapping. The feedback report may include: a HARQ ACK report for a previous transmission of the BS, a CSI report, or both. In some examples, the channel state parameters may include one or more of: BLER, the number of CBs that are negatively acknowledged, the number of CBGs that are NACKed, CQI, SPEF, SNR, LLR quality, parity status, observed best RV, current analog precoder used at the UE, current digital precoder used at the UE, best analog precoder observed at the UE, best digital precoder observed at the UE, observed best DMRS configuration, current interference pattern in frequency and / or time, current interference covariance matrix rank, current interference power, current interference covariance matrix eigenvalues, best resource allocation at the UE, or a combination thereof.

[0110] Operation 1000 may also include the UE receiving, from the BS, configuration of the number of states to be maintained at the UE, channel condition parameters associated with the states, or both. In some examples, the configuration is via RRC signaling or MAC-CE signaling.

[0111] The operation at 1000 may also include: the UE receiving, from the BS, configuration of a quantization level for one or more channel condition parameters among the plurality of channel condition parameters. In some examples, the configuration is via RRC signaling or MAC-CE signaling.

[0112] Operation 1000 may also include: at 1008, the UE receives an indication of one or more actions from the BS in response to sending an indication of the first state. Each of the one or more actions may be indicated separately, or the one or more actions may be indicated together via an index mapped to multiple actions. In some examples, the one or more actions may include one or more of the following for subsequent transmission: MCS, channel rank, transmit precoder at the BS, first analog precoder at the UE, second analog precoder at the BS, first digital precoder at the UE, second digital precoder at the BS, RV, DMRS configuration, resource allocation, or a combination thereof. In some examples, the one or more actions include: one or more actions to be taken by the BS, one or more actions to be taken by the UE, or a combination thereof.

[0113] Fig.11 The invention shows that it may include operations configured to perform the techniques disclosed herein (such as Fig. 9 1100 includes various components (e.g., corresponding to means plus functional components) of the communication device 1100 and the operations shown in FIG. 1100 . The communication device 1100 includes a processing system 1102 coupled to a transceiver 1108 (e.g., a transmitter and / or a receiver). The transceiver 1108 is configured to send and receive signals for the communication device 1100, such as the various signals described herein, via an antenna 1110. The processing system 1102 may be configured to perform processing functions of the communication device 1100, including processing signals received by the communication device 1100 and / or to be transmitted by the communication device 1100.

[0114] The processing system 1102 includes a processor 1104 coupled to a computer-readable medium / memory 1112 via a bus 1106. In certain aspects, the computer-readable medium / memory 1112 is configured to store instructions (e.g., computer-executable code) that, when executed by the processor 1104, cause the processor 1104 to perform Fig. 9, or other operations for performing the various techniques for data transmission configuration using machine learning-based algorithms discussed herein. In certain aspects, the computer-readable medium / memory 1112 stores: code 1114 for receiving a feedback report including an indication of a first state corresponding to a plurality of channel condition parameters from a UE; code 1116 for determining one or more actions based at least in part on the first state; code 1118 for at least signaling to the UE one or more actions to be taken by the UE, etc. In certain aspects, the processor 1104 has circuitry configured to implement the code stored in the computer-readable medium / memory 1112. The processor 1104 includes: circuitry 1124 for receiving a feedback report including an indication of a first state corresponding to a plurality of channel condition parameters from a UE; circuitry 1126 for determining one or more actions based at least in part on the first state; circuitry 1128 for at least signaling to the UE one or more actions to be taken by the UE, etc.

[0115] Fig.12 The invention shows that it may include operations configured to perform the techniques disclosed herein (such as Fig.10 1200 includes a communication device 1200 that includes various components (e.g., corresponding to means plus functional components) of the operations shown in FIG. 1200 . The communication device 1200 includes a processing system 1202 coupled to a transceiver 1208 (e.g., a transmitter and / or a receiver). The transceiver 1208 is configured to send and receive signals for the communication device 1200, such as the various signals described herein, via an antenna 1210. The processing system 1202 may be configured to perform processing functions of the communication device 1200, including processing signals received by the communication device 1200 and / or to be transmitted by the communication device 1200.

[0116] The processing system 1202 includes a processor 1204 coupled to a computer-readable medium / memory 1212 via a bus 1206. In certain aspects, the computer-readable medium / memory 1212 is configured to store instructions (e.g., computer-executable code) that, when executed by the processor 1204, cause the processor 1204 to perform Fig.10, or other operations for performing the various techniques discussed herein for data transmission configuration utilizing machine learning-based algorithms. In certain aspects, the computer-readable medium / memory 1212 stores: code 1214 for receiving signaling from a BS for configuring a mapping of a state index to a channel condition parameter; code 1216 for monitoring transmissions from the BS; code 1218 for sending a feedback report to the BS for the monitored transmission, the feedback report including an indication of a first state index corresponding to a plurality of channel condition parameters based on the mapping; code 1220 for receiving an indication of one or more actions from the BS in response to sending an indication of the first state, and the like. In certain aspects, the processor 1204 has circuitry configured to implement the code stored in the computer-readable medium / memory 1212. Processor 1204 includes: a circuit 1224 for receiving signaling from a BS for configuring a mapping of a state index to a channel state parameter; a circuit 1226 for monitoring transmissions from the BS; a circuit 1228 for sending a feedback report to the BS regarding the monitored transmissions; the feedback report includes an indication of a first state index corresponding to a plurality of channel state parameters based on the mapping; a circuit 1230 for receiving an indication of one or more actions from the BS in response to sending an indication of the first state, and the like.

[0117] Example aspects

[0118] In addition to the above aspects, these aspects can also be combined. Some specific combinations of the aspects are detailed below:

[0119] Aspect 1: A method for wireless communication by a base station (BS), comprising: receiving a feedback report from a user equipment (UE) including an indication of a first state corresponding to multiple channel condition parameters; and determining one or more actions based at least in part on the first state.

[0120] Aspect 2: A method according to Aspect 1, wherein determining the one or more actions based at least in part on the first state includes: determining the one or more actions using a machine learning algorithm using a second state, wherein the second state is based at least in part on the first state.

[0121] Aspect 3: The method according to any one of Aspects 1 or 2 further includes: configuring a mapping of state indexes to channel condition parameters for the UE, wherein the indication of the first state includes one of the state indexes.

[0122] Aspect 4: The method according to any one of aspects 1-3 further comprises: configuring, for the UE, a number of states to be maintained at the UE, channel condition parameters associated with the states, or both.

[0123] Aspect 5: A method according to any one of aspects 1-4, wherein the feedback report includes a hybrid automatic repeat request (HARQ) acknowledgement (ACK) report, a channel state information (CSI) report, or both for a previous transmission by the BS.

[0124] Aspect 6: A method according to any one of Aspects 1-5, wherein the multiple channel condition parameters include one or more of the following: block error rate (BLER), the number of negatively acknowledged code blocks (CB), the number of NACKed CB groups (CBG), channel quality indicator (CQI), spectrum efficiency (SPEF), signal-to-noise ratio (SNR), log-likelihood ratio (LLR) quality, parity check status, observed best redundancy version (RV), current analog precoder used at the UE, current digital precoder used at the UE, best analog precoder observed at the UE, best digital precoder observed at the UE, observed best demodulation reference signal (DMRS) configuration, current interference pattern in frequency and / or time, current interference covariance matrix rank, current interference power, current interference covariance matrix eigenvalues, optimal resource allocation at the UE, or a combination thereof.

[0125] Aspect 7: The method according to Aspect 6 further includes: configuring, for the UE, a quantization level for one or more channel condition parameters among the multiple channel condition parameters.

[0126] Aspect 8: The method according to any one of aspects 1-7, wherein the configuration is via radio resource control (RRC) signaling or medium access control (MAC) control element (CE) signaling.

[0127] Aspect 9: The method according to any one of aspects 2-8, further comprising: determining the second state by adding the first state to a third state maintained at the BS.

[0128] Aspect 10: A method according to Aspect 9, wherein the third state corresponds to one or more of the following: current resource allocation, a digital precoder at the BS, an analog precoder at the BS, a current modulation and coding scheme (MCS), a current data rank, a current action, a current demodulation reference signal (DMRS) configuration, a current transmit power, or a combination thereof.

[0129] Aspect 11: A method according to any one of Aspects 2-10, wherein using a machine learning algorithm to determine the one or more actions includes determining one or more of the following for subsequent transmission: a modulation and coding scheme (MCS), a channel rank, a transmit precoder at a BS, a first analog precoder at a UE, a second analog precoder at the BS, a first digital precoder at the UE, a second digital precoder at the BS, a redundancy version (RV), a demodulation reference signal (DMRS) configuration, resource allocation, or a combination thereof.

[0130] Aspect 12: A method according to any one of Aspects 2-11, wherein the machine learning algorithm comprises a reinforcement learning algorithm, a deep learning algorithm, a supervised learning algorithm, an unsupervised learning algorithm, a Q-learning algorithm, a value reinforcement algorithm, a polarity reinforcement algorithm, or a combination thereof.

[0131] Aspect 13: A method according to any one of Aspects 2-12, wherein the machine learning algorithm is modeled as a Markov decision process (MDP).

[0132] Aspect 14: A method according to any one of Aspects 2-13, wherein the machine learning algorithm is implemented by an artificial neural network.

[0133] Aspect 15: A method according to Aspect 14, wherein: the artificial neural network includes a deep Q network (DQN), the deep Q network includes one or more deep neural networks (DNNs); and using the machine learning to determine the one or more actions includes: passing the one or more second states and one or more action parameters through the one or more DNNs; outputting a value for each action parameter; and selecting the action associated with the optimal output value.

[0134] Aspect 16: A method according to any one of Aspects 2-15, wherein the machine learning uses the states associated with the multiple channel conditions, reward parameters, and action parameters associated with one or more parameters associated with subsequent transmissions, and the reward parameters are associated with the following: signal throughput, spectral efficiency, minimum block error rate (BLER), minimum delay, minimum number of retransmissions, target BLER associated with ultra-reliable low-latency communication (URLLC), target delay associated with URLLC, or a combination thereof.

[0135] Aspect 17: The method according to any one of aspects 1-16, wherein the one or more actions include: one or more actions to be taken by a BS, one or more actions to be taken by a UE, or a combination thereof.

[0136] Aspect 18: The method according to any one of aspects 1-17 further comprises: at least signaling to the UE one or more actions to be taken by the UE.

[0137] Aspect 19: The method according to Aspect 18, wherein the signaling comprises: signaling each of the one or more actions separately; or signaling the one or more actions together via an index mapped to multiple actions.

[0138] Aspect 20: A method according to Aspect 19, wherein: the method also includes sending radio resource control (RRC) signaling or medium access control (MAC) control element (CE) signaling indicating the mapping; and the signaling includes a DCI containing the index.

[0139] Aspect 21: A method for wireless communication by a user equipment (UE), comprising: receiving a configuration of a mapping of a state index to a channel state parameter from a base station (BS); monitoring transmissions from the BS; and sending a feedback report for the monitored transmission to the BS, the feedback report comprising an indication of a first state index corresponding to a plurality of channel state parameters based on the mapping.

[0140] Aspect 22: The method according to aspect 21 further includes: receiving, from the BS, configuration of the number of states to be maintained at the UE, channel condition parameters associated with the states, or both.

[0141] Aspect 23: A method according to any one of Aspects 21 or 22, wherein the feedback report includes a hybrid automatic repeat request (HARQ) acknowledgment (ACK) report, a channel state information (CSI) report, or both for a previous transmission by the BS.

[0142] Aspect 24: A method according to any one of Aspects 21-23, wherein the channel condition parameters include one or more of the following: block error rate (BLER), the number of negatively acknowledged code blocks (CB), the number of negatively acknowledged CB groups (CBG), a channel quality indicator (CQI), a spectral efficiency (SPEF), a signal-to-noise ratio (SNR), a log-likelihood ratio (LLR) quality, a parity check status, an observed best redundancy version (RV), a current analog precoder used at the UE, a current digital precoder used at the UE, a best analog precoder observed at the UE, a best digital precoder observed at the UE, an observed best demodulation reference signal (DMRS) configuration, a current interference pattern in frequency and / or time, a current interference covariance matrix rank, a current interference power, a current interference covariance matrix eigenvalue, an optimal resource allocation at the UE, or a combination thereof.

[0143] Aspect 25: The method according to any one of aspects 21-24 further includes: receiving, from a BS, a configuration of a quantization level for one or more channel condition parameters of the plurality of channel condition parameters.

[0144] Aspect 26: The method according to any one of claims 21, 22 or 25, wherein the configuration is via radio resource control (RRC) signaling or medium access control (MAC) control element (CE) signaling.

[0145] Aspect 27: The method according to any one of aspects 21-26, further comprising: in response to sending the indication of the first state, receiving an indication of one or more actions from the BS.

[0146] Aspect 28: A method according to Aspect 27, wherein the one or more actions include one or more of the following for subsequent transmission: a modulation and coding scheme (MCS), a channel rank, a transmit precoder at a BS, a first analog precoder at a UE, a second analog precoder at the BS, a first digital precoder at the UE, a second digital precoder at the BS, a redundancy version (RV), a demodulation reference signal (DMRS) configuration, resource allocation, or a combination thereof.

[0147] Aspect 29: The method according to aspect 27, wherein the one or more actions include: one or more actions to be taken by the BS, one or more actions to be taken by the UE, or a combination thereof.

[0148] Aspect 30: The method according to Aspect 27, wherein: each of the one or more actions is indicated separately; or the one or more actions are indicated together via an index mapped to multiple actions.

[0149] Aspect 31: An apparatus comprising means for performing the method according to any one of aspects 1 to 30.

[0150] Aspect 31: An apparatus comprising at least one processor and a memory coupled to the at least one processor, the memory comprising code executable by the at least one processor to cause the apparatus to perform the method according to any one of aspects 1 to 30.

[0151] Aspect 32: A computer-readable medium having stored thereon computer-executable code for wireless communication, the computer-executable code, when executed by at least one processor, causes an apparatus to perform a method according to any one of aspects 1 to 30.

[0152] The techniques described herein may be used for various wireless communication technologies, such as NR (e.g., 5G NR), 3GPP Long Term Evolution (LTE), LTE-Advanced (LTE-A), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single Carrier Frequency Division Multiple Access (SC-FDMA), Time Division Synchronous Code Division Multiple Access (TD-SCDMA), and other networks. The terms "network" and "system" are often used interchangeably. A CDMA network may implement radio technologies such as Universal Terrestrial Radio Access (UTRA), cdma2000, and the like. UTRA includes Wideband CDMA (WCDMA) and other variants of CDMA. cdma2000 covers IS-2000, IS-95, and IS-856 standards. A TDMA network may implement radio technologies such as Global System for Mobile Communications (GSM). OFDMA networks may implement radio technologies such as NR (e.g., 5G RA), Evolved UTRA (E-UTRA), Ultra Mobile Broadband (UMB), IEEE 802.11 (WiFi), IEEE 802.16 (WiMAX), IEEE 802.20, Flash-OFDMA, etc. UTRA and E-UTRA are part of Universal Mobile Telecommunications System (UMTS). LTE and LTE-A are versions of UMTS that use EUTRA. UTRA, E-UTRA, UMTS, LTE, LTE-A, and GSM are described in documents from an organization named "3rd Generation Partnership Project" (3GPP). cdma2000 and UMB are described in documents from an organization named "3rd Generation Partnership Project 2" (3GPP2). NR is an emerging wireless communication technology under development.

[0153] In 3GPP, the term "cell" can refer to the coverage area of ​​a Node B (NB) and / or a NB subsystem serving the coverage area, depending on the context in which the term is used. In NR systems, the terms "cell" and BS, next-generation Node B (gNB or gNodeB), access point (AP), distributed unit (DU), carrier, or transmit receive point (TRP) can be used interchangeably. The BS can provide communication coverage for macro cells, pico cells, femto cells, and / or other types of cells. A macro cell can cover a relatively large geographic area (e.g., a radius of several kilometers) and can allow unrestricted access by UEs with service subscriptions. A pico cell can cover a relatively small geographic area and can allow unrestricted access by UEs with service subscriptions. A femto cell can cover a relatively small geographic area (e.g., a home) and can allow restricted access by UEs associated with a femto cell (e.g., UEs in a closed subscriber group (CSG), UEs of users in a home, etc.). A BS for a macro cell can be referred to as a macro BS. A BS for a pico cell can be referred to as a pico BS. A BS for a femto cell may be called a femto BS or a home BS.

[0154] UE may also be referred to as a mobile station, a terminal, an access terminal, a subscriber unit, a station, a customer premises equipment (CPE), a cellular phone, a smart phone, a personal digital assistant (PDA), a wireless modem, a wireless communication device, a handheld device, a laptop computer, a cordless phone, a wireless local loop (WLL) station, a tablet computer, a camera, a gaming device, a netbook, a smartbook, an ultrabook, an appliance, a medical device or medical equipment, a biometric sensor / device, a wearable device such as a smart watch, smart clothing, smart glasses, a smart wristband, smart jewelry (e.g., a smart ring, a smart bracelet, etc.), an entertainment device (e.g., a music device, a video device, a satellite radio device, etc.), a vehicle component or sensor, a smart meter / sensor, an industrial manufacturing equipment, a global positioning system device, or any other suitable device configured to communicate via a wireless or wired medium. Some UEs may be considered machine type communication (MTC) devices or evolved MTC (eMTC) devices. MTC and eMTC UEs include, for example, robots, drones, remote devices, sensors, meters, monitors, location tags, etc. that can communicate with a BS, another device (e.g., a remote device), or some other entity. A wireless node can provide, for example, a connection to or to a network (e.g., a wide area network such as the Internet or a cellular network) via a wired or wireless communication link. Some UEs may be considered Internet of Things (IoT) devices, which may be narrowband IoT (NB-IoT) devices.

[0155] In some examples, access to the air interface can be scheduled. The scheduling entity (e.g., BS) allocates resources for communication between some or all devices and apparatuses within its service area or cell. The scheduling entity may be responsible for scheduling, assigning, reconfiguring, and releasing resources for one or more subordinate entities. That is, for scheduled communications, the subordinate entity utilizes the resources allocated by the scheduling entity. The base station is not the only entity that can be used as a scheduling entity. In some examples, the UE can be used as a scheduling entity and can schedule resources for one or more subordinate entities (e.g., one or more other UEs), and other UEs can utilize the resources scheduled by the UE for wireless communication. In some examples, the UE can be used as a scheduling entity in a peer-to-peer (P2P) network and / or a mesh network. In the mesh network example, in addition to communicating with the scheduling entity, the UEs can also communicate directly with each other.

[0156] The method disclosed herein includes one or more steps or actions for implementing the method. The method steps and / or actions may be interchangeable with each other without departing from the scope of the claims. That is, unless a specific order of steps or actions is specified, the order and / or use of specific steps and / or actions may be modified without departing from the scope of the claims.

[0157] As used herein, a phrase referring to "at least one" of a list of items refers to any combination of those items, including single members. As an example, "at least one of a, b, or c" is intended to cover a, b, c, ab, ac, bc, and abc, as well as any combination with multiple of the same elements (e.g., aa, aaa, aab, aac, abb, acc, bb, bbb, bbc, cc, and ccc or any other order of a, b, and c).

[0158] As used herein, the term "determine" encompasses a wide variety of actions. For example, "determine" may include calculating, computing, processing, deriving, investigating, searching (e.g., searching in a table, a database, or another data structure), ascertaining, and the like. Furthermore, "determine" may include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory), and the like. Furthermore, "determine" may include resolving, selecting, choosing, establishing, and the like.

[0159] The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects. Therefore, the claims are not intended to be limited to the various aspects shown herein, but should be given the full scope consistent with the language of the claims, wherein, unless otherwise specified, references to elements in the singular form are not intended to mean "one and only one", but "one or more". Unless otherwise specifically stated, the term "some" refers to one or more. All structural and functional equivalents of the elements of the various aspects described throughout the present disclosure that are known or will be known to a person of ordinary skill in the art are expressly incorporated herein by reference and are intended to be covered by the claims. In addition, nothing disclosed herein is intended to be dedicated to the public, regardless of whether such disclosure is explicitly recorded in the claims. Any claim element should not be interpreted according to the provisions of 35 U.S.C. § 112 (f), unless the element is explicitly recorded using the phrase "unit for...", or in the case of a method claim, the element is recorded using the phrase "step for...".

[0160] The various operations of the above methods may be performed by any suitable unit capable of performing the corresponding functions. These units may include various hardware and / or software components and / or modules, including but not limited to circuits, digital signal processors (DSPs), application specific integrated circuits (ASICs), or processors (e.g., general or specially programmed processors). Generally, where there are operations shown in the accompanying drawings, those operations may have corresponding corresponding units plus functional components.

[0161] The various illustrative logical blocks, modules, and circuits described in conjunction with the present disclosure may be implemented or performed with a general purpose processor, a DSP, an ASIC, a field programmable gate array (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in an alternative, the processor may be any commercially available processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, a combination of one or more microprocessors and a DSP core, or any other such configuration.

[0162] If implemented in hardware, an example hardware configuration may include a processing system in a wireless node. The processing system may be implemented using a bus architecture. Depending on the specific application of the processing system and the overall design constraints, the bus may include any number of interconnecting buses and bridges. The bus may link together various circuits including a processor, a machine-readable medium, and a bus interface. The bus interface may be used to connect a network adapter, etc., to the processing system via the bus. The network adapter may be used to implement the signal processing functions of the PHY layer. In a user terminal (see Figure 1 ), a user interface (e.g., keyboard, display, mouse, joystick, etc.) may also be connected to the bus. The bus may also link various other circuits, such as timing sources, peripherals, voltage regulators, power management circuits, etc., which are well known in the art and therefore will not be described further. The processor may be implemented with one or more general and / or special purpose processors. Examples include microprocessors, microcontrollers, DSP processors, and other circuits that can execute software. Those skilled in the art will recognize how to best implement the described functionality of the processing system depending on the specific application and the overall design constraints imposed on the entire system.

[0163] If implemented in software, the functions may be stored on or transmitted through a computer-readable medium as one or more instructions or codes. Whether referred to as software, firmware, middleware, microcode, hardware description language or other terms, software should be broadly interpreted as meaning instructions, data or any combination thereof. Computer-readable media include both computer storage media and communication media, including any media that facilitates the transmission of a computer program from one place to another. The processor may be responsible for managing the bus and general processing, including executing software modules stored on a machine-readable storage medium. A computer-readable storage medium may be coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. In an alternative, the storage medium may be integrated into the processor. As an example, a machine-readable medium may include a transmission line, a carrier modulated by data, and / or a computer-readable storage medium having instructions stored thereon separated from a wireless node, all of which may be accessed by a processor through a bus interface. Alternatively or additionally, a machine-readable medium or any part thereof may be integrated into a processor, such as a cache and / or a general register file. As an example, examples of machine-readable storage media may include RAM (random access memory), flash memory, ROM (read only memory), PROM (programmable read only memory), EPROM (erasable programmable read only memory), EEPROM (electrically erasable programmable read only memory), registers, magnetic disks, optical disks, hard drives, or any other suitable storage media, or any combination thereof. The machine-readable medium may be embodied in a computer program product.

[0164] A software module may include a single instruction or many instructions, and may be distributed over several different code segments, between different programs, and over multiple storage media. A computer-readable medium may include multiple software modules. A software module includes instructions that cause a processing system to perform various functions when executed by a device such as a processor. A software module may include a transmission module and a reception module. Each software module may reside in a single storage device or may be distributed over multiple storage devices. For example, when a triggering event occurs, a software module may be loaded from a hard drive into a RAM. During the execution of a software module, a processor may load some of the instructions into a cache to increase access speed. One or more cache lines may then be loaded into a general register file for execution by the processor. When referring to the functions of the following software modules, it will be understood that such functions are implemented by the processor when executing instructions from the software module.

[0165] Additionally, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared (IR), radio and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio and microwave are included in the definition of medium. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Optical disks, where magnetic disks typically reproduce data magnetically, and optical disks use lasers to reproduce data optically. Thus, in some aspects, computer-readable media may include non-transitory computer-readable media (e.g., tangible media). Additionally, for other aspects, computer-readable media may include transitory computer-readable media (e.g., signals). Combinations of the above should also be included within the scope of computer-readable media.

[0166] Thus, certain aspects may include a computer program product for performing the operations presented herein. For example, such a computer program product may include a computer-readable medium having instructions stored (and / or encoded) thereon, which instructions may be executed by one or more processors to perform the operations described herein. For example, for performing the operations described herein and Fig. 9 and / or Fig.10 The operations shown in the instructions.

[0167] In addition, it should be understood that the modules and / or other appropriate units for performing the methods and techniques described herein can be downloaded and / or otherwise obtained by the user terminal and / or base station where applicable. For example, such a device can be coupled to a server to facilitate the transfer of units for performing the methods described herein. Alternatively, the various methods described herein can be provided via a storage unit (e.g., RAM, ROM, a physical storage medium such as a compact disc (CD) or a floppy disk, etc.), so that the user terminal and / or base station can obtain the various methods when the storage unit is coupled to or provided to the device. In addition, any other appropriate technology for providing the methods and techniques described herein to the device can be used.

[0168] It is to be understood that the claims are not limited to the precise configuration and components described above. Various modifications, changes and variations may be made in the arrangement, operation and details of the methods and apparatus described above without departing from the scope of the claims.

Claims

1. A device for wireless communication, comprising: at least one processor; as well as a memory coupled to the at least one processor, the memory comprising code executable by the at least one processor to cause the apparatus to: receiving a feedback report including a first state index from a user equipment (UE), wherein the first state index is mapped to a plurality of channel condition parameters; and determining one or more actions based at least in part on the first state index, Wherein, the code executable by the at least one processor to cause the device to determine the one or more actions based at least in part on the first state index includes code executable by the at least one processor to cause the device to perform the following operations: determine the one or more actions using a machine learning algorithm using a second state, wherein the second state is at least in part based on the first state index.

2. The device according to claim 1, wherein: The memory also includes code executable by the at least one processor to cause the apparatus to: A mapping of state indexes including the first state index to channel condition parameters is configured for the UE.

3. The device according to claim 2, wherein: The configuration is via Radio Resource Control (RRC) signaling or Medium Access Control (MAC) Control Element (CE) signaling.

4. The device according to claim 1, wherein: The memory also includes code executable by the at least one processor to cause the apparatus to: The UE is configured with a number of states to be maintained at the UE, channel condition parameters associated with the states, or both.

5. The device according to claim 1, wherein: The feedback report includes a hybrid automatic repeat request (HARQ) acknowledgement (ACK) report, a channel state information (CSI) report, or both for a previous transmission by the apparatus.

6. The device according to claim 1, wherein: The multiple channel condition parameters include one or more of the following: block error rate (BLER), the number of code blocks (CB) that are negatively acknowledged (NACK), the number of CB groups (CBG) that are NACKed, a channel quality indicator (CQI), a spectral efficiency (SPEF), a signal-to-noise ratio (SNR), a log-likelihood ratio (LLR) quality, a parity check status, an observed best redundancy version (RV), a current analog precoder used at the UE, a current digital precoder used at the UE, a best analog precoder observed at the UE, a best digital precoder observed at the UE, an observed best demodulation reference signal (DMRS) configuration, a current interference pattern in frequency, a current interference pattern in time, a current interference covariance matrix rank, a current interference power, one or more current interference covariance matrix eigenvalues, an optimal resource allocation at the UE, or a combination thereof.

7. The device according to claim 1, wherein: The memory also includes code executable by the at least one processor to cause the apparatus to: One or more quantization levels for one or more channel condition parameters of the plurality of channel condition parameters are configured for the UE.

8. The device according to claim 1, wherein: The memory also includes code executable by the at least one processor to cause the apparatus to: The second state is determined by adding a first state corresponding to the first state index to a third state maintained at the network node.

9. The device according to claim 8, wherein: The third state corresponds to one or more of: current resource allocation, a digital precoder at the device, an analog precoder at the device, a current modulation and coding scheme (MCS), a current data rank, a current action, a current demodulation reference signal (DMRS) configuration, a current transmit power, or a combination thereof.

10. The device according to claim 1, wherein: The code executable by the at least one processor to cause the device to utilize a machine learning algorithm to determine the one or more actions includes code executable by the at least one processor to cause the device to determine one or more of the following for subsequent transmission: a modulation and coding scheme (MCS), a channel rank, a transmit precoder at the device, a first analog precoder at the UE, a second analog precoder at the device, a first digital precoder at the UE, a second digital precoder at the device, a redundancy version (RV), a demodulation reference signal (DMRS) configuration, a resource allocation, or a combination thereof.

11. The device according to claim 1, wherein: The code that can be executed by the at least one processor to enable the device to use a machine learning algorithm to determine the one or more actions includes code that can be executed by the at least one processor to enable the device to perform the following operations: determine the one or more actions using a reinforcement learning algorithm, a deep learning algorithm, a supervised learning algorithm, an unsupervised learning algorithm, a Q-learning algorithm, a value reinforcement algorithm, a polarity reinforcement algorithm, or a combination thereof.

12. The device according to claim 1, wherein: The machine learning algorithm is modeled as a Markov Decision Process (MDP).

13. The device according to claim 1, wherein: The machine learning algorithm is implemented by an artificial neural network.

14. The device according to claim 13, wherein: The artificial neural network comprises a deep Q network (DQN), wherein the deep Q network comprises one or more deep neural networks (DNN); as well as The code executable by the at least one processor to cause the apparatus to use the machine learning algorithm to determine the one or more actions includes code executable by the at least one processor to cause the apparatus to: passing the second state and one or more action parameters through the one or more DNNs; outputting the value for each action parameter; and Select the action associated with the best output value.

15. The device according to claim 1, wherein: The code executable by the at least one processor to cause the device to utilize the machine learning algorithm to determine the one or more actions includes code executable by the at least one processor to cause the device to perform the following operations: input the second state into a machine learning model to predict one or more actions that maximize or minimize a reward parameter, wherein the reward parameter is associated with one or more of the following: signal throughput, spectral efficiency, minimum block error rate (BLER), minimum delay, minimum number of retransmissions, a target BLER associated with ultra-reliable low-latency communication (URLLC), a target delay associated with URLLC, or a combination thereof, and wherein the one or more actions include selecting one or more parameters for subsequent transmissions.

16. The device according to claim 1, wherein The memory also includes code executable by the at least one processor to cause the apparatus to: At least one action among the one or more actions to be taken by the UE is signaled to the UE.

17. The device according to claim 16, wherein: The code executable by the at least one processor to cause the apparatus to signal to the UE the at least one action among the one or more actions to be taken by the UE includes code executable by the at least one processor to cause the apparatus to: signaling each of the one or more actions separately; or Multiple actions are signaled together via an index mapped to the multiple actions.

18. A method for wireless communication by a network node, comprising: receiving a feedback report including a first state index from a user equipment (UE), wherein the first state index is mapped to a plurality of channel condition parameters; and determining one or more actions based at least in part on the first state index, Wherein, determining one or more actions based at least in part on the first state index includes: determining the one or more actions using a machine learning algorithm using a second state, wherein the second state is based at least in part on the first state index.

19. The method according to claim 18, further comprising: The second state is determined by adding a first state corresponding to the first state index to a third state maintained at the network node.

20. The method according to claim 19, wherein: The third state corresponds to one or more of: current resource allocation, a digital precoder at the network node, an analog precoder at the network node, a current modulation and coding scheme (MCS), a current data rank, a current action, a current demodulation reference signal (DMRS) configuration, a current transmit power, or a combination thereof.

21. The method according to claim 18, wherein: Utilizing a machine learning algorithm to determine the one or more actions includes: Determine one or more of the following for subsequent transmission: a modulation and coding scheme (MCS), a channel rank, a transmit precoder at the network node, a first analog precoder at the UE, a second analog precoder at the network node, a first digital precoder at the UE, a second digital precoder at the network node, a redundancy version (RV), a demodulation reference signal (DMRS) configuration, resource allocation, or a combination thereof.

22. The method according to claim 18, wherein: Utilizing a machine learning algorithm to determine the one or more actions includes: The one or more actions are determined using a reinforcement learning algorithm, a deep learning algorithm, a supervised learning algorithm, an unsupervised learning algorithm, a Q-learning algorithm, a value reinforcement algorithm, a polarity reinforcement algorithm, or a combination thereof.

23. The method according to claim 18, wherein: The machine learning algorithm is modeled as a Markov Decision Process (MDP).

24. The method according to claim 18, wherein: The machine learning algorithm is implemented by an artificial neural network.

25. The method of claim 24, wherein: The artificial neural network comprises a deep Q network (DQN), wherein the deep Q network comprises one or more deep neural networks (DNN); as well as Utilizing the machine learning algorithm to determine the one or more actions includes: passing the second state and one or more action parameters through the one or more DNNs; outputting the value for each action parameter; and Select the action associated with the best output value.

26. The method of claim 18, wherein: Utilizing the machine learning algorithm to determine the one or more actions includes: inputting the second state into a machine learning model to predict one or more actions that maximize or minimize a reward parameter, The reward parameter is associated with one or more of the following: signal throughput, spectral efficiency, minimum block error rate (BLER), minimum delay, minimum number of retransmissions, target BLER associated with ultra-reliable low-latency communication (URLLC), target delay associated with URLLC, or a combination thereof, and The one or more actions include selecting one or more parameters for subsequent transmission.

27. The method according to claim 26, wherein: Maximizing or minimizing a reward parameter includes producing a maximum objective function or a minimum objective function.

28. The method of claim 18, further comprising: At least one action among the one or more actions to be taken by the UE is signaled to the UE.

29. The method according to claim 28, wherein: Signaling the UE to the at least one action includes: signaling each of the one or more actions separately; or Multiple actions are signaled together via an index mapped to the multiple actions.

Images