Qualcomm Incorporated

Other

Alcatel Lucent

Taesang Yoo

Xiaoxia Zhang

Tao Luo

Jing X Sun

Siddhartha Mallik

Juan Montojo

Yongbin Wei

Aleksandar Damnjanovic

Naga Bhushan

Tingfang Ji

June Namgoong

Krishna Kiran Mukkavilli

Hamed Pezeshki

Junyi Li

Arumugam Chendamarai Kannan

Peter S Gaal

Durga Prasad Malladi

Srinivas Yerramalli

Sony Akkarakaran

Tamer Adel Kadous

Jay Kumar Sundararajan

Pavan Kumar Vitthaladevuni

Rajat Prakash

Jelena Damnjanovic

Zhifei Fan

Wooseok Nam

Wanshi Chen

Seyed Ali Akbar Fakoorian

Mahmoud Taherzadeh Boroujeni

Madhavan Srinivasan Vajapeyam

Jingchao Bao

Alexandros Manolakos

Hwan Joon Kwon

Hao Xu

Kapil Bhattad

Yi Huang

Simone Merlin

Jun Wang

Aziz Gholmieh

Xiliang Luo

Gokul Sridharan

Gavin Bernard Horn

Xipeng Zhu

Hung Dinh Ly

Joseph Binamira Soriaga

Yisheng Xue

John Edward Smee

Shimman Arvind Patel

Ahmed Kamel Sadek

Kapil Gulati

Jing Lei

Andrei Dragos Radulescu

Byoung-Hoon Kim

Wei Jennifer Yang

Renqiu Wang

Hua Wang

Weimin Duan

Yeliz Tokgoz

Seyedkianoush Hosseini

Sudhir Kumar Baghel

Muhammad Sayed Khairy Abdelghaffar

Shailesh Patil

Makesh Pravin John Wilson

Parag Arun Agashe

Anantharaman Balasubramanian

Qiang Wu

Vinay Chande

Sungwoo Park

Ozcan Ozturk

Yu Zhang

Tianyang Bai

Heechoon Lee

Bilal Sadiq

Masato Kitazoe

Seyong Park

Miguel Griot

Farhad Meshkati

Danlu Zhang

Osok Song

Alberto Rico Alvarino

Alexei Yurievitch Gorokhov

Jung Ho Ryu

Peter Pui Lok Ang

Bin Tian

Hemanth Sampath

Le Liu

Sameer Vermani

Avneesh Agrawal

Yang Yang

Guttorm Ringstad Opshaug

Chenxi Hao

Liangming Wu

Ashok Mantravadi

Dan Vassilovski

Gene Wesley Marsh

Michael Mao Wang

Qiaoyu Li

Joseph Patrick Burke

Disclosed are systems and techniques for wireless communications. For instance, a user equipment (UE) can perform federated learning to generate a first set of updated model parameters corresponding to a machine learning model. In some cases, the UE can receive a request for the first set of updated model parameters from a network entity, wherein the request includes a resource allocation associated with an uplink channel. In some examples, the UE can determine a signal phase corresponding to the uplink channel. In some aspects, the UE can transmit, based on the signal phase, the first set of updated model parameters using the resource allocation on the uplink channel.

Signal synchronization for over-the-air aggregation in a federated learning framework

Systems and techniques are described herein for wireless communication. For example, a computing device (e.g., a user equipment or component thereof) can obtain channel estimation of at least one channel for communication with a network device. The computing device can obtain a location parameter associated with the computing device. The computing device can determine, based at least on the channel estimation, one or more large scale parameters (LSPs). The computing device can further generate an LSP vector associated with the one or more LSPs. The computing device can transmit, to the network device, the LSP vector and the location parameter of the computing device.

Generation of complete large scale parameter maps

An apparatus for wireless communication at a UE is provided. The apparatus is configured to receive a configuration to report interference measurement information indicating interference measurements for each interference measurement resource of a set of interference measurement resources and Rx beam information used by the UE for performing interference measurements on each interference measurement resource of the set of interference measurement resources. The apparatus is configured to receive a set of interference measurement reference signals on the set of interference measurement resources, and to measure interference on each interference measurement resource of the set of interference measurement resources to obtain the interference measurement information. Each interference measurement is through one Rx beam of a set of Rx beams. The apparatus is configured to transmit, in response to the received set of interference reference signals on the set of interference measurement resources, the interference measurement information, and corresponding Rx beam information.

Interference data collection with beam information for ML-based interference prediction

Various aspects of the present disclosure relate to wireless communication. In some aspects, a client may receive an observed environmental vector feedback configuration associated with a reporting procedure for reporting updates corresponding to at least one observed environmental vector that is based at least in part on one or more features associated with an environment of the client, wherein the observed environmental vector comprises input for a first machine learning component that determines client specific parameters for use by a second machine learning component. The client may transmit an update corresponding to the at least one observed environmental vector based at least in part on the observed environmental vector feedback configuration. Numerous other aspects are provided.

Observed environmental vector feedback for wireless communication

A UE may identify a plurality of local model update elements associated with an updated local machine learning model. The updated local machine learning model may have been generated based on a global machine learning model received from a base station and a local dataset. The UE may identify one or more local model update elements of the plurality of local model update elements for update element dropping based on at least one of a channel gain, a PAPR specification, an RF emission specification, or an RF condition. The UE may transmit, to the base station over a multiple access channel via analog signaling, at least some local model update elements of the plurality of local model update elements based on dropping of the identified one or more local model update elements.

Gradient dropping for federated learning

Disclosed are techniques for wireless communication. In an aspect, a first network node determines at least one positioning measurement based on a channel estimate of a positioning reference signal (PRS) transmitted by a second network node, and obtains measurement error feedback for at least the at least one positioning measurement, wherein the measurement error feedback is based on an expected positioning measurement corresponding to the at least one positioning measurement, and wherein the expected positioning measurement is based on a location of the first network node and a location of the second network node.

Measurement error feedback to enable machine learning-based positioning

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may measure one or more linear combinations, of a set of beams, from a radio access network (RAN) node. Accordingly, the UE may transmit a report, based at least in part on measuring the one or more linear combinations, to the RAN node. The one or more linear combinations may be indicated in one or more rules stored in a memory of the UE or may be indicated in a system information block or a radio resource configuration message. Numerous other aspects are described.

Measurements of linear combinations of beams

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

New radio (NR) transmission opportunity (TxOP) structure having flexible starting points are disclosed for wireless communications. When a typical listen before talk (LBT) procedure is performed, the transmitting node will not know ahead of time when the LBT will pass and, thus, when data transmissions can start. NR systems may include a mini-slot design for accommodating transmission units smaller than a slot. Aspects of the present disclosure provide for flexibly increasing the number of potential starting transmission boundaries depending on when the LBT pass is detected. Alternative aspects of the present disclosure determine the potential starting transmission points using a mini-slot based design, a floating slot based design, or a punctured slot based design with a code block group (CBG) level retransmission mechanism.

NR-unlicensed transmission opportunity structure with flexible starting point

Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive an indication of one or more encoding operations to use for encoding a compressed dataset, the one or more encoding operations including a differential encoding operation, or an entropy encoding operation, or both. In some examples, using a neural network, the UE may first encode a dataset based on an additional encoding operation to generate a compressed dataset and then quantize the compressed dataset encoded based on the additional encoding operation. Subsequently, after the dataset has been initially encoded and then quantized, the UE may use the indication of the one or more encoding operations to further encode and compress the dataset. The UE may then transmit the dataset to a second device based on the one or more encoding operations.

Encoding techniques for neural network architectures

Methods, systems, and devices for wireless communications are described. In some examples, a wireless communications system may support machine learning and may configure a user equipment (UE) for machine learning. The UE may transmit, to a base station, a request message that includes an indication of a machine learning model or a neural network function based at least in part on a trigger event. In response to the request message, the base station may transmit a machine learning model, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function and may transmit an activation message to the UE to implement the machine learning model and the neural network function.

Configuring a user equipment for machine learning

Aspects presented herein may enhance the accuracy and/or latency of UE positioning based on crowd-sourcing, where a network entity may compute a position estimate of a UE based on neighbor-cell scan data from the UE and one or more reference UEs. In one aspect, a network entity receives a first set of measurements associated with at least one cell from a UE. The network entity performs a position estimation of the UE based on at least one of the first set of measurements associated with the at least one cell, a second set of measurements for each of a set of reference UEs, or a location of each of the set of reference UEs via an ML model, where the UE and the set of reference UEs include at least one common cell.

Similarity learning for crowd-sourced positioning

Disclosed are techniques for wireless positioning. In an aspect, a base station determines a channel profile for a multipath channel between the base station and a user equipment (UE) based on at least one positioning reference signal transmitted to the UE or received from the UE by the base station on one or more radio beams, compresses the channel profile into a compressed representation of the channel profile, and transmits the compressed representation of the channel profile to a network entity. The network entity receives the compressed representation of the time-angle channel profile and determines a location of the UE based on the compressed representation of the channel profile.

Base station-to-server signaling of time-angle channel profile

Techniques are provide for neural network based positioning of a mobile device. An example method for measuring a channel in a wireless communication system includes: receiving reference signal information; determining one or more channel frequency responses based on the reference signal information and one or more timing hypotheses; determining one or more channel impulse responses comprising a channel impulse response for each of the one or more channel frequency responses; processing the one or more channel impulse responses with a neural network; and determining an output of the neural network.

Neural network based line of sight detection and angle estimation for positioning

Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive a first indication of a first number of antenna ports for which the UE may report channel state information (CSI), and a second indication of a second number of antenna ports on which the UE may measure CSI reference signals (CSI-RSs). The second number may be less than the first number. The UE may receive a third indication of one or more neural networks to be used by the UE for determination of the CSI associated with the first number. The UE may determine the CSI using the one or more neural networks and using measurements made by the UE on the second number as inputs to the one or more neural networks. The UE may transmit a report including the CSI associated with the first number determined via the one or more neural networks.

Techniques for determining channel state information using a neural network model

Growing community of inventors

San Diego, CA, United States of America

Taesang Yoo