Qualcomm Incorporated

Weiliang Zeng

Yisheng Xue

Xiaoxia Zhang

Taesang Yoo

Yongbin Wei

Sanaz Barghi

Siddhartha Mallik

Aleksandar Damnjanovic

Naga Bhushan

Tamer Adel Kadous

Pouriya Sadeghi

Supratik Bhattacharjee

Jing X Sun

Tingfang Ji

Joseph Binamira Soriaga

Krishna Kiran Mukkavilli

June Namgoong

Gautham Hariharan

Navin Dunichand Anwani

Juan Montojo

Alexandros Manolakos

Seyed Ali Akbar Fakoorian

Yu Zhang

Zhifei Fan

Ahmed Kamel Sadek

Pavan Kumar Vitthaladevuni

Hyojin Lee

Qingsi Wang

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.

Gain scaling of input to neural network for end-to-end learning in wireless communication system

A method of wireless communications by a user equipment (UE) includes decoding information received from a base station via a physical downlink shared channel (PDSCH). The UE determines whether to update channel state feedback based on the decoded information. The channel state feedback is updated based on the determination to generate updated channel state feedback. The updated channel state feedback is transmitted to the base station. The UE is configured to update the channel state feedback without an additional measurement of a channel state information reference signal (CSI-RS).

Enhanced user equipment channel state feedback capability with frequent reporting

Various aspects of the present disclosure generally relate to neural network based channel state information (CSI) feedback. In some aspects, a device may obtain a CSI instance for a channel, determine a neural network model including a CSI encoder and a CSI decoder, and train the neural network model based at least in part on encoding the CSI instance into encoded CSI, decoding the encoded CSI into decoded CSI, and computing and minimizing a loss function by comparing the CSI instance and the decoded CSI. The device may obtain one or more encoder weights and one or more decoder weights based at least in part on training the neural network model. Numerous other aspects are provided.

Neural network based channel state information feedback

Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive a set of reference signals from a base station and may determine a first set of channel state parameters. The UE may determine one or more preprocessed decoder parameters based on a downlink channel decoder of the UE and may perform one or more signal processing operations to determine one or more adjustment values for the first set of channel state parameters. The UE may transmit an indication of a the one or more adjustment values to the base station. In some implementations, the UE may transmit an indication of the one or more preprocessed decoder parameters to the base station and the base station may perform the one or more signal processing operations to determine the one or more adjustment values.

Online adjustment for channel state information

Certain aspects of the present disclosure provide techniques for qualifying machine learning model-based channel state information (CSI) predictions. An example method generally includes receiving, from a network entity, a channel state information (CSI) prediction model for quantized CSI, calculating CSI based on downlink reference signal measurements, generating a quantized CSI difference value based a quantization of a difference between the calculated CSI and CSI predicted based on a CSI prediction model, and reporting, to the network entity, the calculated CSI and the quantized CSI difference value.

Qualifying machine learning-based CSI prediction

Methods, systems, and devices for wireless communications are described. A transmitter may identify a contention window for a no-energy-detection protocol for determining a channel availability for a data transmission. After the contention window expires, the transmitter may send a request message to a receiver. The receiver may measure a quality of the request message against a request threshold and transmit a response message if the measured quality exceeds the request threshold. The transmitter may then measure a quality of the response message against a response threshold. If the measured quality exceeds the response threshold, the transmitter may initiate a data transmission with the receiver. If the quality falls below the response threshold or the response message is not received, the transmitter may adjust the contention window and reattempt the no-energy-detection protocol. The contention window may also be adjusted based on an acknowledgement message received after the data transmission.

Channel availability protocol in a shared spectrum

Systems and methods facilitating feedback of robust channel state information (CSI), such as to provide full CSI feedback or otherwise providing CSI feedback, are described. CSI encoders and/or decoders used by network nodes may implement channel compression/reconstruction based upon neural-network (NN) training of collected channels. A structured payload having an interpretable payload portion and an uninterpretable payload portion may utilized with respect to CSI feedback. The channel compression provided according to some aspects of the disclosure supports feedback of robust CSI, in some instances including full CSI, as determined by a particular network node. Other aspects and features are also claimed and described.

Channel state information feedback using channel compression and reconstruction

Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive, from a base station, a request that the UE report a predicted future channel state information measurement. Upon receiving the request, the UE may measure, during a first time period, a channel to obtain measured channel state information. The UE may then determine the predicted future channel state information for a second time period based on the measured channel state information. In some case, the second time period may be subsequent to the first time period. Upon determining the predicted future channel state information measurement, the UE may transmit, to the base station, a channel state information report. In some aspects, the channel state information report may be indicative of the predicted future channel state information measurement.

Forward-looking channel state information prediction and reporting

Two-phase floating-instant coordinated multipoint (CoMP) operation is disclosed. The CoMP operation may be initiated by the network via a CoMP cluster of network nodes or by user equipments (UEs) configured for uplink CoMP transmissions. The first phase is initiated after a first node to conduct a successful full listen before talk (LBT) procedure signals the other participating nodes to perform an abbreviated LBT procedure and identifies the beginning of the second phase. After success of the full and abbreviated LBT procedures, each associated node will initiate a first transmission in the first phase. The leading node determines how many of the other nodes are available for transmission and decides, based on that amount and the rules of the network, whether to continue with the CoMP operations. If enough of the other nodes are available, the participating nodes will conduct the CoMP transmissions at the identified beginning of the second phase.

Floating-instant coordinated multipoint for new radio-unlicensed

Multiple modulation and coding scheme (MCS) tables for new radio (NR) networks is disclosed. Multiple MCS tables are defined with different sets of MCS values selected for improved performance in various transmission conditions. The multiple MCS tables may be known to base stations and user equipments (UEs) in the network. A serving base station may determine a set of transmission characteristics related to a transmission environment of a UE. The base station selects one of the MCS tables of the multiple MCS tables based on the set of transmission characteristics. The selected MCS table includes MCS values that may provide improved performance for the UE considering one or more of the characteristics of the set of transmission characteristics. The base station will then transmit an indication of the selected MCS table to the served UE.

Multiple MCS tables for NR

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may transmit, to a base station, a report indicating that a cooperative reception group including the UE and one or more other UEs connected to the UE over an out-of-band link supports one or more downlink transmit beamforming techniques. The UE may determine, based at least in part on the one or more downlink transmit beamforming techniques supported by the cooperative reception group, a downlink transmit precoder associated with relaxed beamforming nulling. The UE may decode a downlink transmission received from the base station based at least in part on the downlink transmit precoder and one or more decoded bits received from the one or more other UEs in the cooperative reception group. Numerous other aspects are provided.

UE cooperative reception and cooperative transmission for quality of service demanding applications

A method for wireless communications, comprising: signaling, to a user equipment (UE), an indication of one of at least two rules regarding how quasi co-location (QCL) configured for the UE should be applied for one or more DMRS ports; and sending downlink transmission to the UE with the one or more DMRS ports; receiving signaling from a base station and processing signals received on one or more DMRS ports based on the indicated rule. Said method improves communications between access points and stations in a wireless network.

Determining DMRS average delay and delay spread under smooth pre-coding

Methods, systems, and devices for wireless communications are described, and relate to a base station to communicating with a user equipment (UE) over a channel. A first device (for example, the base station or the UE) may use a trained neural network to estimate one or more link performance metrics associated with the channel. Predicting the link level performance may include determining one or more neural network weights associated with one or more input parameters associated with the channel to estimate the one or more link performance metrics. The first device may report feedback to the second device based on the estimated link performance metrics. Based on the feedback, the second device may adapt the link by adjusting channel parameters to improve the reliability or efficacy of later transmissions.

Neural-network-based link-level performance prediction

Methods, systems, and devices for wireless communication are described. A base station may multiplex reference signal (RS) transmissions for different antenna ports over a set of resource elements using a code division multiplexing (CDM) group spreading in a time dimension and a frequency dimension. The base station may transmit the RS transmissions over the set of resource elements based at least in part on the multiplexing to a user equipment (UE). The UE may demultiplex the RS transmissions based at least in part on a CDM group and may select at least one of the time dimension or the frequency dimension over which to demultiplex the RS transmissions based on a determination or other information.

Unified reference signal design for enhanced component carriers

A per-transmit time interval (TTI) rank control within a transmission opportunity (TxOP) is described for downlink coordinated multiple point (CoMP) operations. In operation, a base station may determine rank and data rate for communication during a transmission opportunity (TxOP) at the beginning of the TxOP. Each TxOP may include a plurality of transmission time intervals (TTIs). For each TTI of the TxOP, the base station computes one or more channel quality indicator (CQI) back-offs using one or more outer loop adaptation processes per TTI, and updates rank and the data rate using the CQI back-offs. Once updated, the base station transmits the downlink data during each TTI according to the updated rank and the data rate.

Growing community of inventors

San Diego, CA, United States of America

Weiliang Zeng