US 11,818,648 B2
Apparatus, method and computer program for routing data in a dual or multi-connectivity configuration
Udaya Raju Chowta, Ulm (DE); Bertold Dickhaus, Olpe (DE); Fahad Syed Muhammad, Orsay (FR); Afef Feki, Sceaux (FR); and Veronique Capdevielle, Magny les Hameaux (FR)
Assigned to Nokia Solutions and Networks Oy, Espoo (FI)
Filed by Nokia Solutions and Networks Oy, Espoo (FI)
Filed on Mar. 17, 2021, as Appl. No. 17/203,875.
Claims priority of application No. 20164790 (EP), filed on Mar. 23, 2020.
Prior Publication US 2021/0297928 A1, Sep. 23, 2021
Int. Cl. H04W 40/32 (2009.01); H04W 76/15 (2018.01); G06N 20/00 (2019.01); H04W 80/02 (2009.01); H04W 84/18 (2009.01); H04W 92/20 (2009.01)
CPC H04W 40/32 (2013.01) [G06N 20/00 (2019.01); H04W 76/15 (2018.02); H04W 80/02 (2013.01); H04W 84/18 (2013.01); H04W 92/20 (2013.01)] 24 Claims
OG exemplary drawing
 
1. An apparatus comprising:
at least one processor; and
at least one non-transitory memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least perform at a central unit of a master node:
receiving an individual packet data unit from one or more incoming packet data units, wherein each packet data unit is to be transmitted to a user equipment device via one amongst respective paths towards at least one distributed unit of the master node and at least one secondary node, the user equipment device being in dual or multi-connectivity with the master node and the at least one secondary node;
collecting a respective data set for each of the paths;
receiving, from a machine learning model inputting each data set, at least one first estimated delay of transmission for at least one first path comprising the at least one distributed unit, and at least one second estimated delay of transmission for at least one second path comprising the at least one secondary node;
selecting, from each of the paths, a path based on at least the at least one first estimated delay of transmission for the at least one first path comprising the at least one distributed unit, and the at least one second estimated delay of transmission for the at least one second path comprising the at least one secondary node;
transmitting the packet data unit to the user equipment device via the selected path;
receiving a plurality of measured delays of transmission for the selected path at a first frequency;
obtaining network data for the selected path at a second frequency;
wherein the first frequency is higher than the second frequency; and
training the machine learning model at least partially using the plurality of measured delays of transmission and the network data;
wherein to train the machine learning model, at least some of the network data obtained at the second frequency is duplicated during a time interval to correspond to respectively different measured delays of transmission of the plurality of measured delays of transmission received at the first frequency.