US 11,811,527 B2
Detecting control information communicated in frame using a neural network
Jean-Luc Olivés, Espoo (FI); Vesa Starck, Helsinki (FI); Mikko Tapio Kokkonen, Helsinki (FI); and Gerardo Moreno Crespo, Espoo (FI)
Assigned to NOKIA TECHNOLOGIES OY, Espoo (FI)
Appl. No. 17/420,303
Filed by NOKIA TECHNOLOGIES OY, Espoo (FI)
PCT Filed Jan. 7, 2019, PCT No. PCT/EP2019/050263
§ 371(c)(1), (2) Date Jul. 1, 2021,
PCT Pub. No. WO2020/143902, PCT Pub. Date Jul. 16, 2020.
Prior Publication US 2022/0094464 A1, Mar. 24, 2022
Int. Cl. H04L 27/06 (2006.01); H04L 1/00 (2006.01); G06N 3/08 (2023.01); H04L 1/1829 (2023.01); H04W 76/28 (2018.01)
CPC H04L 1/005 (2013.01) [G06N 3/08 (2013.01); H04L 1/0026 (2013.01); H04L 1/0036 (2013.01); H04L 1/1864 (2013.01); H04W 76/28 (2018.02)] 20 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
at least one processor; and
at least one memory including computer program code, wherein the at least one memory and computer program code are configured to, with the at least one processor, to cause the apparatus to perform:
acquiring, from a received frame, a set of samples associated with a position of a control information element in the frame;
inputting the set of samples to nodes of an input layer of a neural network, the input layer having the same number of nodes as a number of samples in the set of samples;
processing the set of samples in the neural network that has been trained, before receiving the set of samples, to decode one or more determined values of the control information element and to detect discontinuous transmission;
outputting, in an output layer of the neural network, an indicator indicating a decoded value of the control information element comprised in the set of samples or an indicator indicating the discontinuous transmission,
wherein the neural network is trained by training inputs to the neural network, the training inputs comprising possible values of the control information element as corrupted by simulated noise, wherein the simulated noise simulates one or more radio channel models, the training inputs further comprising payload data samples or samples comprising randomly generated bits as a discontinuous transmission training input.