Nokia Technologies Oy

Nokia Solutions and Networks Oy

Sajad Rezaie

Oana-Elena Barbu

Keeth Saliya Jayasinghe Laddu

Amir Mehdi Ahmadian Tehrani

Example embodiments provide improved spatial beam predictions based on multiple assistance information parameters. An apparatus may be configured to configure a plurality of subspaces and corresponding machine learning models based on a dataset comprising multiple assistance information parameters, select a subset of the subspaces for inference by the machine learning models based on available assistance information, and determine a preferred spatial beam prediction based on the inferences. Apparatuses, methods, and computer programs are disclosed.

Apparatuses and methods for spatial beam prediction with multiple assistance information

A device for reconstructing a channel state information (C) and a method of reconstructing a channel state information, wherein the device for reconstructing includes circuitry for reconstructing the channel state information depending (C) on a first encoding (T) of the channel state information (C) understandable by a first decoder, and determining the first encoding (T) depending on a second encoding (W, . . . , W) of the channel state information (C) understandable by a second decoder. A device for encoding a channel state information and a method of encoding a channel state information, wherein the device for encoding includes circuitry for determining a first encoding of the channel state information understandable by an encoder, determining a second encoding of the channel state information understandable by a decoder depending on the first encoding.

Method and device

An apparatus comprising circuitry configured to: receive a plurality of signal samples, wherein at least two of the signal samples are received on different carrier frequencies or over different durations; convert the plurality of signal samples to a common data format that stores as entries the samples, a stored entry corresponding to a sample measured with a carrier frequency, a time, and a receive antenna; generate entries of the common data format that are missing due to a sample not being measured with a carrier frequency, time, and receive antenna; wherein the entries of the common data format are ranked, wherein a higher ranking entry is considered more relevant than a relatively lower ranking entry; and generate at least one positioning measurement with a machine learning model, based on the entries of the common data format and the ranking of the entries of the common data format.

Growing community of inventors

Aalborg, Denmark

Sajad Rezaie