Abstract: Methods, systems, and devices for wireless communications are described. Generally, the described techniques at a user equipment (UE) provide for efficiently reporting channel state information (CSI) to a base station with an appropriate level of accuracy. In particular, the base station may indicate a level of accuracy to the UE for reporting CSI. The UE may encode the CSI using a first neural network, and the base station may decode the CSI using a second neural network. The first and second neural networks may form a neural network pair, and the UE may train the neural network pair based on the level of accuracy indicated by the base station. For example, the base station may indicate a loss function corresponding to a level of accuracy with which CSI is to be reported by the UE, and the UE may train the neural network pair using the loss function.

Abstract: Wireless communications systems may configure a subset of allocated resources (e.g., one or more resource elements (REs) of one or more allocated resource blocks (RBs)) as peak reduction tones (PRTs) for a peak-cancelation signal. For instance, wireless communications systems may configure a fixed PRT allocation based on a Costas array. In some examples, each column of a Costas array may correspond to a RB of a set of allocated resources. A transmitting device may thus identify one or more PRT REs based on the Costas array and a mapping of allocated RBs to the columns of the Costas array. For instance, a transmitting device may identify a pattern of PRT REs to use for a peak-cancellation signal based at least in part on a configured Costas array (e.g., where the peak-cancellation signal may reduce peaks of a corresponding data signal to ultimately reduce peak-to-average power ratio (PAPR) of a transmission).

Abstract: Wireless communications systems and methods related to determining a QCL for receiving a reference signal are provided. A user equipment (UE) determines a QCL configuration for receiving a reference signal during a first time period based on a transmit beam direction to be used for transmission during the first time period. The user equipment receives the reference signal using a first receive beam direction based on the QCL configuration while transmitting a first communication signal using the transmit beam direction in a common frequency band during the first time period.

Abstract: Certain aspects of the present disclosure provide techniques for generating and decoding orthogonal frequency division (OFDM) waveforms with peak reduction tones (PRTs) designed to reduce PAPR. By generating PRT tones with a machine learning (e.g., neural network) based encoder and mapping some of the PRT tones to subcarriers used for physical channels or signals, PAPR may be reduced while efficiently using system resources.

Abstract: Methods, systems, and devices for wireless communications are described. A base station may associate a first and second port with a set of antenna elements based on a first and second linear precoding vector. The base station may generate coefficients indicating a first combination and a second combination of a first data set for a first user equipment (UE) and a second data set for a second UE. The base station may apply the first and second linear precoding vectors to the first and second combinations, respectively, transmit a first demodulation reference signal (DMRS) and a second DMRS using a first comb corresponding to the first and second ports, and transmit at least a third DMRS indicating the coefficients using a second comb. The UEs may extract data from the precoded combinations by applying the coefficients and estimating channels associated with the first and second DMRS.

Abstract: A method of wireless communication by a user equipment (UE) receives a channel state information (CSI) decoder and CSI encoder from a base station on a physical downlink control channel (PDSCH) or a media access control-control element (MAC-CE). The method also includes training the CSI decoder and CSI encoder based on observed channel and interference conditions to obtain updated decoder coefficients and updated encoder coefficients. The method further includes receiving an indication of resources for transmission of the updated encoder coefficients and updated decoder coefficients. The method includes transmitting the updated decoder coefficients and updated encoder coefficients to the base station in accordance with the indication of resources. Further, the method includes receiving an updated CSI decoder and updated CSI encoder from the base station for further training.

Abstract: Bandwidth part (BWP) configurations supporting various communication approaches (e.g., full duplex and/or half duplex operations) are described. Full duplex (FD) frequency-based BWP configurations may, for example, be configured as a subset of defined BWP resources for supporting full duplex operation by base stations and/or user equipments (UEs). Usable bandwidth of a FD frequency-based BWP configuration may be selected from half duplex frequency-based BWPs in legacy BWPs. Bandwidths of usable BWPs for a FD frequency-based BWP configuration may be selected so as to be non-overlapping in frequency. Transition between configurations and modes (e.g., between full duplex frequency-based BWP configurations, between half duplex and full duplex modes, etc.) may be managed to avoid periods in which a communication device cannot perform any uplink or downlink transmissions due to switching between defined BWP configurations, or otherwise reduces BWP switching time.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may receive a transformer configuration that includes a transmitter neural network configured to be used to generate at least one latent vector corresponding to one or more computation tasks of a plurality of computation tasks associated with a transformer-based cross-node machine learning system. The UE may transmit the at least one latent vector based at least in part on instantiating the transmitter neural network. Numerous other aspects are described.

Abstract: A method of wireless communication by a user equipment (UE) includes receiving different sets of parameters from different sources as input to a receiver neural network. The method also includes receiving, from a base station, a set of target long-term energy values associated with the receiver neural network. The method further includes calculating a scaling factor for each of the different sets of parameters based on the set of target long-term energy values, and separately scaling each of the different sets of parameters based on the scaling factor calculated for that set in order to generate multiple sets of scaled parameters. The method still further includes transmitting the multiple sets of scaled parameters to the receiver neural network.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may interleave a signal, that is to be transmitted using discrete Fourier transform (DFT) spread orthogonal frequency-division multiplexing (DFT-s-OFDM), and a negative of the signal to obtain an interleaved signal prior to performing a DFT on the interleaved signal. The UE may transmit a DFT-s-OFDM signal that is based at least in part on the interleaved signal. Numerous other aspects are provided.

Abstract: Methods, systems, and devices for wireless communications are described. A base station may associate a first port and a second port each with a set of antenna elements based on a first and a second linear precoding vector, where the phases of the ports are coherent. The base station may generate coefficients indicating a first combination and a second combination of a first data set for a first user equipment (UE) and a second data set for a second UE. In some cases, the base station may apply the first linear precoding vector to the first combination, apply the second linear precoding vector to the second combination, and transmit the first combination using a first transmission beam corresponding to the first port and the second combination using a second transmission beam corresponding to the second port to the first and second UEs.

Abstract: Methods, systems, and devices for wireless communications are described in which a base station may identify a null space matrix that lies within a null space of an effective channel matrix for communications between the base station and a user equipment (UE). An indication of the null space matrix may be provided to the UE, and the null space matrix used to determine modifications to a precoding matrix. The base station and UE may determine a redistribution matrix that provides a reduced variance of transmission powers for a number of transmission channels, where a product of the null space matrix and the redistribution matrix may be computed and added to the precoding matrix to generate a modified precoding matrix. The modified precoding matrix may be used to generate the communications from the base station and UE with reduced power variance across channels.

Abstract: A method of wireless communication, by a user equipment (UE), includes receiving multiple neural network training configurations for channel state feedback (CSF). Each configuration corresponds to a different neural network framework. The method also includes training each of a group of neural network decoder/encoder pairs in accordance with the received training configurations. A method of wireless communication, by a base station, includes transmitting multiple neural network training configurations to a user equipment (UE) for channel state feedback (CSF). Each configuration corresponds to a different neural network framework. The method also includes receiving a neural network decoder/encoder pair trained in accordance with the training configurations.

Abstract: A second base station may allocate a set of downlink resources for a downlink transmission. The set of downlink resources may include a subset of null resources, which may serve as a signature for the second base station. The second base station may transmit downlink data or reference signals on the set of downlink resources including the subset of null resources. A first UE may experience the downlink transmission from the second base station as interference. The first UE may identify the set of downlink resources. The first UE may identify the subset of null resources. The first UE may transmit, to a first base station, an indication of the subset of null resources. The first base station may identify the second base station based on the indication of the subset of null resources. The first base station may transmit, to the second base station, an interference coordination message.

Abstract: Methods, systems, and devices for wireless communications are described. A user equipment (UE) may perform a measurement operation to attain multiple measurements to report to a base station. The measurements may correspond to a first number of bits if reported. The UE may compress the measurements using an encoder neural network (NN) to obtain an encoder output indicating the measurements. This encoder output may include a second number of bits that is less than the first number of bits. The UE may report the encoder output to the base station in this compressed form. At the base station, the encoder output may be decompressed according to a decoder NN. Once the base station decompresses the encoder output, the UE and base station may communicate according to the measurements determined from the decompression. In some cases, the base station may perform load redistribution based on the measurements.

Abstract: Methods, systems, and devices for wireless communication are described. In multi-user (MU) multiple-input multiple-output (MIMO) wireless communications systems, a base station may perform signaling to indicate information related to ports assigned to one or more UEs. The base station may also signal information regarding ports shared with other UEs, which the UE may use when performing channel estimation. In some examples, the base station may signal a number of ports used per sub-band to the UE or a number of demodulation reference signal (DMRS) ports used by the UE. The base station may signal a sub-set of a total number of ports that are shared by other UEs overlapping with the resource allocation of the UE. The signaling may indicate a number of combs used by the base station in the resource allocation for the UE.

Abstract: Aspects of the present disclosure describe receiving, from at least one cell of a zone of multiple cells, a zone-specific signal related to the zone of multiple cells, wherein the zone of multiple cells operate using synchronized timing, acquiring, based on the zone-specific signal, a timing synchronization with a cell of the zone of multiple cells, and communicating with the cell based on the timing synchronization.

Abstract: The present disclosure provides for adaptive resource management in new radio operations that adapts a numerology including a subcarrier spacing and/or cyclic prefix for a user equipment (UE) traveling at a high speed. A base station may transmit via a plurality of remote radio heads (RRH) to a user equipment (UE) is moving along a high speed track. The base station may transmit, in a first time period, using a first numerology including a first subcarrier spacing and a first cyclic prefix ratio, a first transmission for the UE. The base station may transmit, in a subsequent time period, using a second numerology including a second subcarrier spacing and a second cyclic prefix ratio, a second transmission for the UE. At least one of the second subcarrier spacing is different than the first subcarrier spacing or the second cyclic prefix ratio is different than the first cyclic prefix ratio.

Abstract: Techniques for peak-to-average power ratio (PAPR) reduction are described. Wireless devices may use one or more PAPR shaping resources, such as expanded bandwidth and/or pulse-shaping filtering, for shaping a signal to reduce PAPR. For example, expanded bandwidth may be utilized for adding a cyclic affix (CA), such as may comprise a cyclic prefix (CP), cyclic suffix (CS), etc., and combinations thereof, to a frequency domain data signal to provide a CP augmented frequency domain data signal used to generate a reduced PAPR time domain data signal. Additionally or alternatively, pulse-shaping filtering may be applied to a frequency domain signal to provide a pulse-shaped frequency domain data signal used to generate a reduced PAPR time domain data signal. Other aspects and features are also claimed and described.

Abstract: This disclosure provides systems, methods, and apparatus, including computer programs encoded on computer storage media, for wireless communication. In one or more aspects, a user equipment (UE) is configured to transmit capability information of the UE for channel state information (CSI) reporting. The UE is further configured to receive, from a base station, configuration information for the CSI reporting. The configuration information indicates a CSI reference resource including a plurality of a first type of time units for CSI reference signal (CSI RS) monitoring, a CSI target resource including one or more of a second type of time units for which channel state information is to be estimated, or both. The CSI target resource is later in time as compared to the CSI reference resource. Other aspects and features are also claimed and described.