公开(公告)号：WO2020088747A1

公开(公告)日：2020-05-07

申请号：PCT/EP2018/079735

申请日：2018-10-30

Applicant: NOKIA SOLUTIONS AND NETWORKS OY ,  TECHNISCHE UNIVERSITAET MUENCHEN

Inventor： ALI-TOLPPA, Janne Tapio ,  SCHULTZ, Benedek ,  KAJÓ, Márton

IPC: H04L12/26 ,  H04L12/24

Abstract: According to an aspect, there is provided a local diagnosis system comprising means for performing the following. The local diagnosis system detects one or more anomaly events associated with a communications network. Each anomaly event defines an anomaly pattern describing a data point in a performance indicator space. Then, the local diagnosis system updates one or more local cluster models to incorporate the one or more anomaly patterns within complexity constraints. Each of the one or more local cluster models corresponds to a different diagnosis label defining a diagnosis. In response to failing according to one or more pre-defined criteria to incorporate, in the updating, the one or more anomaly patterns to the one or more local cluster models, the local diagnosis system forwards at least the one or more local cluster models and one or more associated diagnosis labels to a central diagnosis system for further diagnosis.

公开(公告)号：WO2022028687A1

公开(公告)日：2022-02-10

申请号：PCT/EP2020/072019

申请日：2020-08-05

Applicant: NOKIA SOLUTIONS AND NETWORKS OY ,  TECHNISCHE UNIVERSITAET MUENCHEN

Inventor： KAJÓ, Márton ,  ALI-TOLPPA, Janne Tapio ,  MWANJE, Stephen

IPC: G06N3/04 ,  G06N3/08 ,  G06N7/00 ,  G06N20/00 ,  H04L12/24 ,  G06N3/0454 ,  G06N3/084 ,  G06N3/088 ,  G06N7/005 ,  H04L41/142 ,  H04L41/16 ,  H04L41/5009 ,  H04L43/08

Abstract: A method for a neural network comprising an input layer, an output layer and one or more intermediate layers is described. The neural network is arranged to generate an output vector of data values at the output layer corresponding to a learned representation of an input vector of data values that is input to the neural network. The method comprises accessing a set of data variables that are determined according to respective entries of output vectors, the output vectors generated on the basis of the evaluation of the neural network on input vectors of data values selected from a training dataset of input vectors, evaluating a predictive model over the set of data variables to determine a subset of data variables and modifying the predictive model and the neural network on the basis of the evaluation whereby the evaluation of the subset of data variables for subsequent input vectors of data values that are input to the neural network generate output vectors of data values that are grouped, according to a measure of similarity, into at least two substantially disjoint subsets.

公开(公告)号：WO2020108747A1

公开(公告)日：2020-06-04

申请号：PCT/EP2018/082814

申请日：2018-11-28

Applicant: NOKIA SOLUTIONS AND NETWORKS OY ,  TECHNISCHE UNIVERSITAET MUENCHEN

Inventor： SCHULTZ, Benedek ,  ALI-TOLPPA, Janne Tapio ,  KAJÓ, Márton

IPC: H04W24/02 ,  H04L12/24

Abstract: Network management apparatus and methods are described. A network management apparatus comprises network data receiving means for receiving network data that is representative of the current condition of a communications network, the network data comprising a plurality of values indicative of the performance of network elements; network data transformation means for transforming the received network data into a network state vector that is indicative of a current state of the network; and network state prediction means for predicting a future network state vector of the network from the current network state vector, the network state prediction means comprising a self- learning prediction module having a memory for storing at least one internal state.

公开(公告)号：WO2022106942A1

公开(公告)日：2022-05-27

申请号：PCT/IB2021/060108

申请日：2021-11-02

Applicant: NOKIA SOLUTIONS AND NETWORKS OY ,  TECHNISCHE UNIVERSITÄT MÜNCHEN

Inventor： KAJÓ, Márton ,  SCHULTZ, Benedek ,  MWANJE, Stephen

IPC: H04L41/14 ,  H04L41/16

Abstract: Apparatuses and methods in a communication system are disclosed. In a network element, an encoder module obtains as an input network data that is representative of the current condition of the communications network, the network data comprising a plurality of values indicative of the performance of network elements and performs (800) feature reduction providing at its output a set of activations. A clustering module performs (802) batch normalisation and an amplitude limitation to the output of the encoder module to obtain normalised activations. A clustering control module calculates a projection of the normalised activations and determines (804) a clustering loss. A decoder module calculates (806) a reconstruction loss. The network element backpropagates the reconstruction loss and the clustering loss through the modules.