Abstract: Disclosed are techniques for environment sensing. In an aspect, a transmitter base station determines a configuration for a radio frequency (RF) sensing signal, the configuration determined based at least in part on coordination among a plurality of base stations, and transmits the RF sensing signal to at least one receiver base station based on the configuration. In an aspect, a receiver base station receives a configuration for an RF sensing signal, the configuration determined based at least in part on coordination among a plurality of base stations, receives, from at least one transmitter base station, the RF sensing signal, and detects at least one target object based, at least in part, on reception of the RF sensing signal.

Abstract: Techniques are provided for utilizing QCL relationships with AI/ML models and reference signals. An example method for providing positioning reference signal configuration information includes receiving, from a wireless node, an indication of a quasi co-location (QCL) relationship between an AI/ML model and a reference signal, configuring one or more positioning reference signal resources based at least in part on the indication of the QCL relationship, and providing configuration information for the one or more positioning reference signal resources to the wireless node.

Abstract: Methods, systems, and devices for wireless communications are described. A first wireless device may be configured to communicate signaling with a second wireless device, an additional wireless device, or both, and perform, based on the signaling, one or more inferences using a machine learning model. The first wireless device may transmit, to the second wireless device, an indication that the machine learning model was applicable for performing the one or more inferences, and receive, from the second wireless device, control signaling indicating that the machine learning model is applicable for communications that are associated with a first set of conditions associated with the communication of the signaling, where the control signaling indicates a model identifier (ID) associated with the machine learning model, that the first set of conditions is associated with the machine learning model, or both.

Abstract: Aspects described herein relate to using machine learning (ML) models for performing channel state information (CSI) encoding or decoding, CSI-reference signal (RS) optimization, channel estimation, etc. The ML models can be trained by a user equipment (UE), separately by the UE and a network node (e.g., base station), or jointly by the UE and network node.

Abstract: A communications transceiver system that employs self-interference mitigation techniques can detect static objects by using techniques that introduce angular or Doppler diversity. This can include moving Tx and Rx antennas and/or performing beam sweeping. When processing RF sensing data from reflected RF signals, self-interference mitigation techniques can be used and compensation can be made for the movement and/or beam sweeping to allow for both self-interference mitigation and detection of static objects.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may receive information to update a machine learning model associated with a training iteration of the machine learning model. The UE may transmit a local loss value based at least in part on the training iteration of the machine learning model within a time window to report the local loss value for the training iteration, the time window having an ending. Numerous other aspects are described.

Abstract: A method, apparatus, or computer-readable medium with instructions for beam management for radio frequency (RF) sensing at a wireless device. The wireless device performs a first beamsweep of an RF signal and measures a reflection of the RF signal based on the first beamsweep. The wireless device performs a second beamsweep of the RF signal, wherein the first beamsweep is based on a different parameter than the second beamsweep and measures the reflection of the RF signal based on the second beamsweep. The wireless device selects a beam for RF sensing based on the first beamsweep and the second beamsweep.

Abstract: Certain aspects of the present disclosure provide techniques for quasi-model (QML) relation indication and configuration for artificial intelligence (AI)/machine learning (ML) air interface operation. An example method, performed at a first wireless node, generally includes transmitting, to a second wireless node, an indication that a first machine learning (ML) model shares one or more properties with at least a second ML model; and utilizing at least the first ML model to communicate with the second wireless node.

Abstract: Techniques are provided herein for signaling path selection modes for positioning. An example method of signaling a path selection mode includes obtaining one or more signal measurement values for a reference signal, determining one or more propagation paths based on the one or more signal measurement values and the path selection mode, and report the one or more propagation paths and the path selection mode to a location server.

Abstract: Methods, systems, and devices for wireless communications are described. A user equipment (UE) may communicate signaling with a network entity, the signaling indicating a first bandwidth size and a second bandwidth size associated with a projection parameter (W1) and a compression parameter (W2), respectively. The second bandwidth size may be less than the first bandwidth size. In some cases, the UE may determine W1 and W2 based on a first machine learning (ML) model and a second ML model, respectively. The UE may project a portion of received channel state information (CSI) associated with the first bandwidth size onto a sub-space defined by W1 and may compress a portion of the projection associated with the second bandwidth size based on W2. The UE may transmit channel state feedback including the projection, the compression of the projection, a compression of W1, a compression of W2, or any combination thereof.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may receive a generative channel model (GCM) that outputs channel information pertaining to a digital twin (DT). The UE may receive a configuration of a downlink reference signal. The UE may perform channel estimation using the configuration of the downlink reference signal. The UE may transmit a report, wherein the report includes a precoding indicator, and wherein at least one of computation of the precoding indicator, or a channel estimation algorithm for the channel estimation, uses the GCM. Numerous other aspects are described.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a network node may receive a configuration corresponding to one or more channel state information (CSI) report settings associated with a reference signal resource set. The network node may transmit at least one CSI report that indicates a primary set of CSI measurements, of a plurality of CSI measurements associated with the reference signal resource set, and a secondary set of CSI measurements of the plurality of CSI measurements, wherein the plurality of CSI measurements comprises at least one of a physical layer reference signal received power (RSRP) value or a physical layer signal-to-interference-plus-noise ratio (SINR), and wherein the primary set of CSI measurements comprises at least one of a strongest RSRP value or a strongest SINR associated with the plurality of CSI measurements. Numerous other aspects are described.

Abstract: Methods, systems, and devices for wireless communications are described. A network entity may transmit, and a user equipment (UE) may receive, signaling identifying a configuration of a set of multiple machine learning (ML) models for channel characteristic prediction. The channel characteristic prediction may include a channel characteristic prediction for each ML model of the set of multiple ML models based on a respective reference signal resource of the set of multiple reference signal resources. The network entity may obtain an input to the set of multiple ML models based on performing one or more measurements associated with the set of multiple reference signal resources. The network entity may output, and the UE may obtain, the input. The UE may process the input using at least one ML model of the set of multiple ML models to obtain the channel characteristic prediction of the at least one ML model.

Abstract: Aspects presented herein may enable a network entity to configure a group of UEs to simultaneously transmit reference signals and to simultaneously transmit gradient vectors to the network entity, such that the network entity may receive the gradient vectors from the group of UEs as an aggregated gradient vector over the air. In one aspect, a base transmits, to a group of UEs, a configuration that configures the group of UEs to simultaneously transmit one or more group-common reference signals and to simultaneously transmit one or more gradient vectors associated with a federated learning procedure. The network entity receives, from the group of UEs, the one or more group-common reference signals and the one or more gradient vectors based on the configuration via multiple channels. The network entity calculates an average gradient vector based on the one or more group-common reference signals and the one or more gradient vectors.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a base station may transmit, to a user equipment (UE), a federated learning configuration that indicates one or more parameters of a federated learning procedure associated with a machine learning component. The base station may receive a local update associated with the machine learning component from the UE based at least in part on the federated learning configuration. Numerous other aspects are provided.

Abstract: Various aspects of the disclosure relate to power control for independent links. For example, power control at a device may be based on transmissions on multiple links. In some aspects, the independent links may involve a first device (e.g., a user equipment) communicating via different independent links with different devices (e.g., transmit receive points (TRPs) or sets of TRPs). For example, the first device may communicate with a second device (e.g., a TRP) via a first link and communicate with a third device (e.g., a TRP) via a second link. In some scenarios, power control for the first device may be based on power control commands received on multiple links. In some scenarios, a power control constraint may be met taking into account the transmission power on multiple links.

Abstract: In one aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus may be a user equipment (UE) or a component thereof. The apparatus may be configured to receive pilot signals from a base station on a first subset of a set of antenna ports of a channel. The apparatus may be further configured to measure a first set of values corresponding to the first subset of the set of antenna ports based on receiving the pilot signals transmitted from the base station on the first subset of the set of antenna ports. The apparatus may be further configured to derive a second set of values corresponding to a second subset of the set of antenna ports of the channel based on receiving the pilot signals on the first subset of the set of antenna ports.

Abstract: Methods, systems, and devices for wireless communications are described. A user equipment (UE) may encode channel state feedback (CSF) information to compress the CSF information to a first encoding output associated with a first dimensional space, and apply entropy coding to the first encoding output of the channel state feedback information. The entropy coding may transform the first encoding output to a second encoding output associated with a second dimensional space that is smaller than the first dimensional space of the first encoding output. The UE may transmit a CSF message comprising the second encoding output. A network device may receive the CSF message and apply entropy decoding to the compressed CSF information to partially decompress the compressed CSF information to a first decoding output. The network device may decode the first decoding output to completely decompress the compressed CSF information to a second decoding output.

Abstract: Certain aspects of the present disclosure provide techniques for channel estimation based on transmission spatial information. An example method of wireless communication by a user equipment includes receiving an indication of transmission spatial information associated with a network entity; receiving a reference signal from the network entity based on the transmission spatial information; and transmitting, to the network entity, channel state information (CSI) based on the received reference signal and the transmission spatial information.

Abstract: Certain aspects of the present disclosure provide techniques for wireless communications by an apparatus. A method includes sending signaling indicative of one or more power ratios, wherein each of the one or more power ratios is based on a transmit power for transmitting sounding reference signal (SRS) using a corresponding antenna and a receive power for receiving channel state information reference signal (CSI-RS) using the corresponding antenna, and wherein the one or more power ratios are associated with one or more antennas; sending a first CSI-RS; receiving a first SRS; receiving first channel state feedback (CSF) corresponding to the CSI-RS; and performing channel estimation based on the first CSF and the first SRS.