Abstract: Disclosed is a method comprising obtaining a plurality of handover parameter values, using a first machine learning model to select a subset of handover parameter values from the plurality of handover parameter values, obtaining historical information of a plurality of terminal devices, determining a first set of optimal handover parameter values for the plurality of terminal devices from the subset of handover parameter values, tagging the first set of optimal handover parameter values with the historical information of the plurality of terminal devices to obtain a labelled dataset, and training a second machine learning model with the labelled dataset, wherein the trained second machine learning model is capable of predicting a second set of optimal handover parameter values for a first terminal device based on historical information of the first terminal device.

Abstract: Disclosed is a method comprising obtaining historical data from a plurality of access nodes comprised in a network, determining, for at least one of the plurality of access nodes, a region comprising a set of values for threshold pairs that comprise minimum and maximum threshold pairs, determining, from the region, one threshold pair, wherein the threshold pair defines pre-determined thresholds for determining if a cell is to be switched on or switched off, providing the threshold pair to the network for deployment, and collecting data regarding at least one key performance indicator from the plurality of access nodes.

Abstract: Disclosed is a method comprising obtaining historical data from a plurality of access nodes comprised in a network, determining, for at least one of the plurality of access nodes, a region comprising a set of values for threshold pairs that comprise minimum and maximum threshold pairs, determining, from the region, one threshold pair, wherein the threshold pair defines pre-determined thresholds for determining if a cell is to be switched on or switched off, providing the threshold pair to the network for deployment, and collecting data regarding at least one key performance indicator from the plurality of access nodes.

Abstract: An apparatus for anomaly detection in a network, using an autoencoder including an encoder and a decoder.

Abstract: Systems, methods, apparatuses, and computer program products for determining a grid-of-beams (GoB) are provided. One method may include collecting network data for training a neural network, train the neural network, using the collected data, to learn a non-discounted cumulative reward Q that evaluates a benefit of including a given beam into a grid-of-beams (GoB), iteratively applying the trained neural network to select at least one optimal beam to include in the grid-of-beams (GoB), and selecting one or more beams from the grid-of-beams (GoB) to transmit to a user equipment or to receive transmission from the user equipment.

Abstract: The invention relates to a method for configuring carrier aggregation for a current user equipment (14) located in a wireless communication network in a coverage area of a base station which is capable of emitting a plurality of carrier components with a plurality transmission beams per carrier component, the method comprising: retrieving quality metrics beams from a database comprising quality metrics associated with a carrier component of the plurality of carrier components and with a user equipment of a plurality of user equipments; determining, using the quality metrics, one or more candidate carrier components to be configured or de-configured for the current user equipment as secondary carrier components; configuring or de-configuring said one or more carrier components.

Abstract: An apparatus for anomaly detection in a network, using an autoencoder including an encoder and a decoder.

Abstract: Systems, methods, apparatuses, and computer program products for determining a grid-of-beams (GoB) are provided. One method may include collecting network data for training a neural network, train the neural network, using the collected data, to learn a non-discounted cumulative reward Q that evaluates a benefit of including a given beam into a grid-of-beams (GoB), iteratively applying the trained neural network to select at least one optimal beam to include in the grid-of-beams (GoB), and selecting one or more beams from the grid-of-beams (GoB) to transmit to a user equipment or to receive transmission from the user equipment.

Abstract: Disclosed is a method comprising obtaining a plurality of handover parameter values, using a first machine learning model to select a subset of handover parameter values from the plurality of handover parameter values, obtaining historical information of a plurality of terminal devices, determining a first set of optimal handover parameter values for the plurality of terminal devices from the subset of handover parameter values, tagging the first set of optimal handover parameter values with the historical information of the plurality of terminal devices to obtain a labelled dataset, and training a second machine learning model with the labelled dataset, wherein the trained second machine learning model is capable of predicting a second set of optimal handover parameter values for a first terminal device based on historical information of the first terminal device.

Abstract: In some embodiments, a computing system maintains, in a memory, information on one or more terminal devices, a dictionary of beams and information on a spatial grid. The computing system maps each terminal device to the spatial grid based on up-to-date results of radio measurements and calculates, for each spatial element, a load caused by the one or more terminal devices based on the mapping. The computing system evaluates, for each combination of a beam and a terminal device, a reference signal received power, RSRP, based on results of the radio measurements. The computing system calculates a first map of expected spatial distribution of traffic based on the calculated loads and beam-specific second maps of expected RSRP based on the values of the RSRP. Finally, the computing system causes performing beamforming optimization based on the first map and second maps.

Abstract: In some embodiments, a computing system maintains, in a memory, information on one or more terminal devices, a dictionary of beams and information on a spatial grid. The computing system maps each terminal device to the spatial grid based on up-to-date results of radio measurements and calculates, for each spatial element, a load caused by the one or more terminal devices based on the mapping. The computing system evaluates, for each combination of a beam and a terminal device, a reference signal received power, RSRP, based on results of the radio measurements. The computing system calculates a first map of expected spatial distribution of traffic based on the calculated loads and beam-specific second maps of expected RSRP based on the values of the RSRP. Finally, the computing system causes performing beamforming optimization based on the first map and second maps.

Abstract: The invention proposes a method for allowing radio communication transitions between packet and circuit switched modes in a cellular radio communication system comprising a core network, a first radio access network supporting a first mode among packet and circuit switched modes and a second radio access network supporting a second mode among packet and circuit switched modes. The first and second radio access networks include base stations covering respective cells, the base stations being capable of communications with mobile stations, an overlap existing between the cells of the first and second radio access networks, the method comprising the step of detecting and conditionally informing a mobile station, when the mobile station is in communication with a base station of the first radio access network covering a cell overlapping with another cell covered by another base station of the second radio access network.