Nokia Technologies Oy

Nokia Solutions and Networks Oy

Nokia Corporation

Mikko Johannes Honkala

Mikko Aleksi Uusitalo

Dani Johannes Korpi

Francesco Cricri

Akos Vetek

Harri Lindholm

Janne Matti Juhani Huttunen

Emre Baris Aksu

Leo Karkkainen

Leo Mikko Johannes Kärkkäinen

Miska Matias Hannuksela

Jani Lainema

Kim Blomqvist

Xingyang Ni

Janne Huttunen

Zexian Li

Ian Justin Oliver

Jakob Hoydis

Athul Prasad

Faycal Ait Aoudia

Juha-Pekka Luoma

Leo Kärkkäinen

Guido Grassel

Asta Maria Karkkainen

Sampo Vesa

Jani Matti Johannes Moilanen

Lauri Ilari Kuru

Caglar Aytekin

Teemu Mikael Veijalainen

Elina Maria Inkeri Ollila

Satu Rajala

Tapio Taipalus

Yan Qing Cui

Michael Woldegebriel

Vesa Starck

Mathias Berglund

Kiti Müller

Leo KÄRKKÄINEN

Leo Karkkaien

Teemu Mikael Veijalaine

In some examples, a node for a telecommunication network includes a neural-network-based receiver for uplink communications. The node is configured to modify the neural-network-based receiver to generate a set of modified receiver frameworks defining respective different versions for the receiver, using each of the modified receiver frameworks, generate respective measures representing bits encoded by a signal received at the node, calculate a value representing a variance of the measures, and on the basis of the value, determine whether to select the signal received at the node for use as part of a training set of data for the neural-network-based receiver.

Neural-network-based receivers

An apparatus, method and SW for HARQ control are disclosed. The method includes: analyzing () a beginning part of a data packet received on a specific frequency band and in a scheduled slot between a user apparatus and a radio access network to predict a success of decoding the data packet after received in full; and controlling () a hybrid automatic repeat request procedure associated with the data packet based on the predicted success using the specific frequency band and the scheduled slot for full-duplex inband signaling.

Apparatus, method and SW for HARQ control

A method of detection of a recurrent feature of interest within a signal including: obtaining evidence, based on a signal, the evidence including a probability density function for each of a plurality of parameters for parameterizing the signal, including at least one probability density function for a parameter, of the plurality of parameters, that positions a feature of interest within signal data of the signal; parameterizing a portion of the signal data from the signal based upon a hypothesis that a point of interest in the signal data is a position of the feature of interest; determining a posterior probability of the hypothesis being true given the portion of the signal data by combining a prior probability of the hypothesis and a conditional probability of observing the portion of the signal data given the hypothesis.

Determining whether a hypothesis concerning a signal is true

The embodiments relate to a method comprising compressing input data (I) by means of at least a neural network (E,); determining a compression rate for data compression; miming the neural network (E,) with the input data (I) to produce an output data (c); removing a number of elements from the output data (c) according to the compression rate to result in a reduced form of the output data (me); and providing the reduced form of the output data (me) and the compression rate to a decoder (D,). The embodiments also relate to a method comprising receiving input data (me) for decompression; decompressing the input data (me) by means of at least a neural network (D,); determining a decompression rate for decompressing the input data (me); miming the neural network (D,) with input data (me) to produce a decompressed output data (ĩ); padding a number of elements to the compressed input data (me) according to the decompression rate to produce an output data (ĩ); and providing the output data (ĩ).

Method, an apparatus and a computer program product for video encoding and video decoding

According to an example embodiment, a radio receiver includes at least one processor and at least one memory including computer program code. The at least one memory and the computer program code may be configured to, with the at least one processor, cause the radio receiver to: obtain a data array including a plurality of elements, wherein each element in the plurality of elements in the data array corresponds to a subcarrier in a plurality of subcarriers and to a timeslot in a time interval; obtain a reference signal array representing a reference signal configuration applied during the time interval; implement a neural network; input data into the neural network, wherein the data includes at least the data array and the reference signal array. A radio receiver, a method and a computer program product are disclosed.

Radio receiver

A method includes maintaining a set of parameters or weights derived through online learning for a neural net; transmitting an update of the parameters or weights to a decoder; deriving a first prediction block based on an output of the neural net using the parameters or weights; deriving a first encoded prediction error block through encoding a difference of the first prediction block and a first input block; encoding the first encoded prediction error block into a bitstream; deriving a reconstructed prediction error block based on the first encoded prediction error block; deriving a second prediction block based on an output of the neural net using the parameters or weights and the reconstructed prediction error block; deriving a second encoded prediction error block through encoding a difference of the second prediction block and a second input block; and encoding the second encoded prediction error block into a bitstream.

Apparatus, a method and a computer program for video coding and decoding

A method, apparatus and computer program, the method comprising: receiving a first biopotential signal obtained by a first capacitive sensor; receiving a second biopotential signal obtained by a second capacitive sensor, the first capacitive sensor and the second capacitive sensor being positioned at different locations on a subject; synchronising biopotential signals obtained by the first capacitive sensor and the second capacitive sensor by applying a time adjustment to biopotential signals obtained by at least one of the first capacitive sensor or the second capacitive sensor; wherein features in at least one of the first biopotential signal and the second biopotential signal are used to synchronise the biopotential signals obtained by the first capacitive sensor and the second capacitive sensor.

Method, apparatus and computer program for synchronizing biopotential signals

This document discloses a solution for detecting interference in a radio access network. According to an aspect, a method includes as performed by a network node of the radio access network: acquiring a first equalized signal representing a signal received by a first radio head serving a terminal device, the first equalized signal including a signal received by the first radio head from the terminal device; acquiring a second equalized signal representing a signal received by a second radio head not serving the terminal device, wherein the second radio head is spatially distant from the first radio head; cross-correlating the first equalized signal with the second equalized signal and determining, on the basis of said cross-correlating, whether or not the second equalized signal also includes a signal received from the terminal device; and as a result of the second equalized signal being determined to include the signal received from the terminal device, causing execution of an interference management action.

Interference detection in radio access network

Disclosed is a method comprising providing a first resource grid as input to a first machine learning algorithm, obtaining a second resource grid as output from the first machine learning algorithm, and transmitting a signal comprising the second resource grid by using orthogonal frequency-division multiplexing modulation.

Machine learning for orthogonal frequency-division multiplexing based communication

A technique includes receiving, from one or more sensors, sensor data samples; receiving radio network information data samples associated with a radio network; determining, based on an association of one or more received sensor data samples with one or more of the received radio network information data samples, a first set of one or more associated sensor and radio network information data samples; developing a model that is trained based on at least a portion of the first set the associated sensor and radio network information data samples that are relevant to performance of the radio network; and improving performance of the radio network based on at least the model.

Radio-network self-optimization based on data from radio network and spatiotemporal sensors

According to an aspect, a method in the radio head comprises receiving a reference signal from a terminal device, and computing a channel estimate on the basis of the reference signal and transferring the channel estimate with a first precision level to the network node over a link between the radio head and the network node. The first precision level defines a first number of bits carrying the channel estimate, then the method determines to change the precision level and, in response to the determining, transfers the channel estimate or a further channel estimate with a second precision level to the network node. The second precision level defines a second number of bits carrying the channel estimate or the further channel estimate, the second number being different from the first number.

Transfer of channel estimate in radio access network

An apparatus comprises: at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to: use a first base station central unit coupled to a first base station distributed unit in dependence on a computational capacity of the first base station central unit.

Apparatus, method and computer program

An approach is provided for synchronizing an impedance cardiography ('ICG') measurement period with an electrocardiography ('ECG') signal to reduce patient () auxiliary current. The approach involves measuring an ECG signal of a patient via an ECG device (). The approach also involves processing the ECG signal to cause, at least in part, a detection of one or more ECG features of the signal. The approach further involves synchronizing a start, a stop, or a combination thereof of a measurement of an ICG signal of the patient via an ICG device () based, at least in part, on the detection of the one or more ECG features. The measurement of the ICG signal includes injecting an electrical current into the patient () for a duration of the measurement.

Method and apparatus for synchronizing impedance cardiography with electrocardiography to lower patient auxiliary current

A method comprising: deriving a first prediction block () at least partly based on an output of a neural net () using a first set of parameters; deriving a first encoded prediction error block (-) through encoding a difference of the first prediction block and a first input block; encoding () the first encoded prediction error block into a bitstream; deriving a first reconstructed prediction error block () from the first encoded prediction error block; deriving a training signal () from one or both of the first encoded prediction error block and/or the first reconstructed prediction error block (); retraining () the neural net () with the training signal () to obtain a second set of parameters for the neural net (); deriving a second prediction block () at least partly based on an output of the neural net using the second set of parameters; deriving a second encoded prediction error block (-) through encoding a difference of the second prediction block and a second input block; and encoding () the second encoded prediction error block into a bitstream. The invention relates to image or video encoding or decoding, especially by online training a neural network () that is in the prediction loop.

An apparatus configured to generate an output quality error estimate using a machine-learning error estimation model to compare an output meeting a predetermined quality threshold with an output image reconstructed from a plurality of images, and provide the output quality error estimate for use in estimating if a second subsequent image is required, in addition to a first subsequent image to obtain a cumulative output having an output quality error meeting a predetermined error threshold. Also an apparatus configured, using a received output quality error estimate generated using a machine-learning error estimation model as above, to estimate if a second subsequent image is required, in addition to a first subsequent image, to obtain a cumulative output having an output quality error meeting a predetermined error threshold.

Growing community of inventors

Espoo, Finland

Mikko Johannes Honkala