公开(公告)号：US20210007120A1

公开(公告)日：2021-01-07

申请号：US16981060

申请日：2018-03-20

Applicant: Nokia Technologies Oy

Inventor： Suresh KALYANASUNDARAM ,  Klaus PEDERSEN ,  Hans KROENER ,  Rajeev AGRAWAL

IPC: H04W72/12 ,  H04W72/08 ,  H04L1/00 ,  H04W72/04 ,  H04W28/04 ,  H04L5/00 ,  H04B7/0452

Abstract: Methods, devices, and computer program products concerning accommodating ultra-reliable low latency communications (URLLC) traffic by making a determination of which mobile broadband (MBB) user equipment (UE) to puncture, where in response to an indication that URLLC traffic needs to be scheduled in the midst of ongoing MBB transmissions in a wireless communications network, the determination is made from a plurality of MBB UEs with the ongoing MBB transmissions, of a set of the plurality scheduled for transmission in a slot required by the URLLC traffic. From that set of MBB UEs, a subset of MBB UEs for puncturing is chosen, where that choice is made at least in relation to accommodating a reliability constraint of the URLLC traffic, maximizing a sum proportional fairness metric of the plurality of MBB UEs, and minimizing a block error rate in a computation of a proportional fairness metric for each UE of the subset.

公开(公告)号：US20230214648A1

公开(公告)日：2023-07-06

申请号：US17928712

申请日：2021-05-26

Applicant: NOKIA TECHNOLOGIES OY

Inventor： Qi LIAO ,  Fahad SYED MUHAMMAD ,  Veronique CAPDEVIELLE ,  Afef FEKI ,  Suresh KALYANASUNDARAM ,  Ilaria MALANCHINI

IPC: G06N3/08 ,  H04B7/06 ,  G06N3/045

CPC classification number: G06N3/08 ,  H04B7/0617 ,  G06N3/045

Abstract: A deep transfer reinforcement learning (DTRL) method based on transfer learning within a deep reinforcement learning (DRL) framework is provided to accelerate the GoB optimization decisions when experiencing environment changes in the same source radio network agent or when being applied from a source radio network agent to a target radio network agent. The transferability of the knowledge embedded in a pre-trained neural network model as a Q-approximator is exploited, and a mechanism to transfer parameters from a source agent to a target agent is provided, where the transferability criterion is based on the similarity measure between the source and target domain.

公开(公告)号：US20210218460A1

公开(公告)日：2021-07-15

申请号：US17252966

申请日：2019-06-20

Applicant: Nokia Technologies Oy

Inventor： Ian Dexter GARCIA ,  Igor FILIPOVICH ,  Chandrasekar SANKARAN ,  Hua XU ,  Suresh KALYANASUNDARAM ,  Jamil SHIHAB ,  Rajeev AGRAWAL ,  Anand BEDEKAR

IPC: H04B7/06 ,  H04B7/0452 ,  G06N20/00

Abstract: Systems, methods, apparatuses, and computer program products for multi-user (MU) multiple-input multiple-output (MIMO) user pairing selection are provided. One method may include selecting multi-user multiple input multiple output (MU MIMO) candidate beams using deep neural network(s) (DNNs), and selecting paired users based on the selected beams. The deep neural network(s) (DNNs) are trained to maximize multi-user priority metric (MU-PM) or a heuristic of the multi-user priority metric (MU-PM).