Abstract: An apparatus, method and computer program is described above: determining a coupling gain for a user device within a cell of a mobile communication system; comparing the determined coupling gain with a coupling gain threshold; setting the user device to operate in a device-specific CSI-RS mode of operation in the event that the determined coupling gain is less than the coupling gain threshold, wherein, in the device-specific CSI-RS mode of operation, the user device is configured to use device-specific reference signal transmissions for the determination of channel state information; and setting the user device to operate in a cell-specific CSI-RS mode of operation in the event that the determined coupling gain is not less than the coupling gain threshold, wherein, in the cell-specific CSI-RS mode of operation, the user device is configured to use cell-specific reference signal transmissions for the determination of channel state information.

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: Embodiments of the present disclosure relate to compressed channel state information transfer. According to embodiments of the present disclosure, the radio unit at the base station receives signals from the terminal device. The radio unit (RU) compresses the signal based on the weights of beams and transmits the compressed signal to the baseband unit (BBU) at the base station. The baseband unit reconstructs the channel based on the compressed signal and the coefficients. In this way, the amount of information conveyed across the RU-BBU interface is reduced, and the accuracy of channel estimation is improved. Further, the sounding reference signal (SRS) channel can be obtained at the BBU with reduced number of SRS transmissions.

Abstract: Methods and apparatus for realizing dynamic point selection improve the cell-edge throughput and geometric mean of release 8/9 UEs. The method comprises a step of receiving estimated channel quality reported by all the TPs in the CoMP set, a step of switching the serving TP for the UEs based on the estimated channel quality and/or cell load, a step of forwarding the scheduled data from the primary TP to the serving TP, a step of separately transmitting PDCCH to the UEs by the primary TP and transmitting PDSCH to the UEs by the serving TP. the implementation of the method and apparatus improves the cell-edge throughput and geometric mean of UE throughput by serving UEs instantaneously from the cell that provides better throughput accounting for fast time-scale channel fluctuations and/or load. It helps in reducing interference due to data getting drained faster when users are served from cells with better channel conditions.

Abstract: According to an aspect, there is provided a computing device for controlling terminal device uplink transmission power. Each terminal device is configured to determine uplink transmission power based on two power control parameters—a target received power for full pathloss compensation and a pathloss compensation coefficient. The computing device initializes a deep Q-learning network in which a state is defined as cell-specific pairs of the power control parameters, an action is defined as a selection of valid values of power control parameters for a cell and a reward is calculated based on the information on data traffic. The computing device trains the deep Q-learning network to approximate a Q value function, determines optimal power control parameters based on thereon and causes transmitting them to access nodes.

Abstract: A solution for allocating N radio beam configurations to access nodes including, in some embodiments, a method comprising storing a beam library defining a set of radio beam configurations for a coverage area. A coverage area is divided into sub-areas, and a set of channel quality metrics is collected, as is a traffic density parameter for each sub-area. The following steps are performed: selecting N radio beam configurations from the set of radio beam configurations; mapping each sub-area with one of the selected N radio beam configurations; computing a service performance for the selected N radio beam configurations; the service performance is weighted by the traffic density parameters. The service performance are stored as linked to each radio beam configuration of the selected N radio beam configurations. Based on the stored service performances, a set of N radio beam configurations is selected and the N radio beam configurations are allocated.

Abstract: A solution for allocating N radio beam configurations to access nodes including, in some embodiments, a method comprising storing a beam library defining a set of radio beam configurations for a coverage area. A coverage area is divided into sub-areas, and a set of channel quality metrics is collected, as is a traffic density parameter for each sub-area. The following steps are performed: selecting N radio beam configurations from the set of radio beam configurations; mapping each sub-area with one of the selected N radio beam configurations; computing a service performance for the selected N radio beam configurations; the service performance is weighted by the traffic density parameters. The service performance are stored as linked to each radio beam configuration of the selected N radio beam configurations. Based on the stored service performances, a set of N radio beam configurations is selected and the N radio beam configurations are allocated.

Abstract: An apparatus is disclosed, comprising means for identifying a plurality of user equipment (UE), each transmitting one or more uplink packets for decoding at a base station associated with a given cell of a radio network. The apparatus further comprises means for clustering the identified user equipment into joint processing groups, each joint processing group comprising the identities of two or more user equipment as clustered and means for performing, in a first processing stage, joint processing of the uplink data streams for identified user equipment within common joint processing groups using one or more first processing algorithms to produce corresponding first processed uplink data streams.

Abstract: According to an example embodiment, a method may include receiving, by a controller from a radio access network (RAN) node within a wireless network, at least one of a neural network support information, and a measurement information that includes one or more measurements by the radio access network node or one or more measurements by a wireless device that is in communication with the radio access network node; determining, by the controller based on the at least one of the neural network support information and the measurement information, a configuration of a neural network for the radio access network node; an sending, by the controller to the radio access network node, neural network configuration information that indicates the configuration of the neural network for the radio access network node.

Abstract: According to an aspect, there is provided a computing device for controlling terminal device uplink transmission power. Each terminal device is configured to determine uplink transmission power based on two power control parameters: a target received power for full pathloss compensation and a pathloss compensation coefficient. The computing device comprises means for performing the following. Information on data traffic in a plurality of cells is maintained in a database. The computing device initializes a deep Q-learning network in which a state is defined as cell-specific pairs of the power control parameters, an action is defined as a selection of valid values of power control parameters for a cell and a reward is calculated based on the information on data traffic. The computing device trains the deep Q-learning network to approximate a Q value function, determines optimal power control parameters based on thereon and causes transmitting them to access nodes.

Abstract: Disclosed is a method for determining open loop power control parameters. A first set of key performance indicators associated with a first set of open loop power control parameters is obtained from one or more first base stations. A statistical model is determined based at least partly on the first set of key performance indicators and the first set of open loop power control parameters. A second set of open loop power control parameters is determined based at least partly on the statistical model.

Abstract: Methods and Systems are disclosed for allocating a time-frequency resource of a large cell base station either to a selected first user equipment associated with the base station, or to a selected self-backhaul link associated with a small cell within the coverage area of the large cell base station. In embodiments, data is provided representing target allocation proportions of the time-frequency resource, for each of the first user equipment(s) and the one or more backhaul link(s). Allocations are made, for each of a series of time units or intervals, either to a selected first user equipment or backhaul link based on the target data.

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.

Abstract: An apparatus is disclosed, comprising means for identifying a plurality of user equipment (UE), each transmitting one or more uplink packets for decoding at a base station associated with a given cell of a radio network. The apparatus further comprises means for clustering the identified user equipment into joint processing groups, each joint processing group comprising the identities of two or more user equipment as clustered and means for performing, in a first processing stage, joint processing of the uplink data streams for identified user equipment within common joint processing groups using one or more first processing algorithms to produce corresponding first processed uplink data streams.

Abstract: This document discloses a solution for selecting a transmission direction in a cell.

Abstract: According to an example embodiment, a method may include receiving, by a controller from a radio access network (RAN) node within a wireless network, at least one of a neural network support information, and a measurement information that includes one or more measurements by the radio access network node or one or more measurements by a wireless device that is in communication with the radio access network node; determining, by the controller based on the at least one of the neural network support information and the measurement information, a configuration of a neural network for the radio access network node; an sending, by the controller to the radio access network node, neural network configuration information that indicates the configuration of the neural network for the radio access network node.

Abstract: An apparatus including an input configured to receive at least one notification from at least one radio access network node controller, the at least one notification caused to notify the apparatus of at least one physical channel assignment policy indication; an assigner configured to implement at least one physical channel assignment based on the at least one physical channel assignment policy indication; an output configured to generate and transmit at least one assignment notification to at least one user equipment associated with the apparatus, the assignment notification caused to notify the at least one user equipment of the at least one physical channel assignment.

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: A base station including a memory and a processor. The memory is configured to store computer readable instructions. The processor is configured to execute the computer readable instructions such that the memory, the processor and the computer readable instructions cause the base station to order a plurality of reception devices according to an amount of transmission resources required to transmit transmission data to each reception device, assign transmission resources in a time slot in blocks to each of the plurality of reception devices in order from a first reception device requiring the least amount of transmission resources to a reception device among the plurality of reception devices requiring a greatest amount of transmission resources, the time slot being divided into a plurality of symbols, and transmit the transmission data to the plurality of reception devices using the assigned transmission resources.

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).