Selection of power control parameters

1. An apparatus, comprising:

at least one processor; and

at least one memory including computer program code,

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to perform:

determine, from a given set of radio access network parameters, a selection of radio access network parameters which have an effect on uplink power control;

train a neural network to determine uplink power control parameters, utilizing as an input the selection of radio access network parameter;

utilize the trained neural network, with as an input the selection of radio access network parameters, obtain as an output a set of initial uplink power control parameters;

when training the neural network, select a performance metric comprising at least one of the following: user equipment normalized mean data rate, normalized cell edge rate and latency; and

select to the set of initial uplink power control parameters those parameters which give highest value to the selected performance metric.

2. The apparatus of claim 1 , wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus further to perform:

utilize as an input in the training of the neural network at least one of the following: one or more radio access network configuration, varying radio access network conditions, network inter site distance, number of users per cell.

3. The apparatus of claim 1 , wherein the uplink power control parameters comprise target received power P 0 , and a fractional pathloss compensation factor α.

4. An apparatus, comprising:

at least one processor; and

at least one memory including computer program code,

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to perform:

determine, from a given set of radio access network parameters, a selection of radio access network parameters which have an effect on uplink power control;

train a neural network to determine uplink power control parameters, utilizing as an input the selection of radio access network parameter;

utilize the trained neural network, with as an input the selection of radio access network parameters, obtain as an output a set of initial uplink power control parameters,

wherein given set of radio access network parameters comprise base station positions, estimated path loss exponent, cell-specific loads, inter site distance, average number of users per cell, network center frequency and average packet size.

5. A network element in a radio access network, comprising:

at least one processor; and

at least one memory including computer program code,

the at least one memory and the computer program code configured to, with the at least one processor, cause the network element at least to:

receive a set of initial uplink power control parameters provided by a trained neural network;

utilize the set of initial uplink power control parameters in an operating radio access network and select from the received set the parameters giving best performance as the uplink power control parameters to be applied in the radio access network wherein when utilizing the set of initial uplink power control parameters in an operating radio access network, apply each uplink power control parameter of the set a predetermined amount of time; and

when utilizing the set of initial uplink power control parameters in an operating radio access network, monitor the performance of the network, and if the performance deteriorates, reject the current uplink power control parameters and apply utilize of the different parameters from the set of initial uplink power control parameters.

6. A method in a radio access network, comprising:

determining, from a given set of radio access network parameters, a selection of radio access network parameters which have an effect on uplink power control;

training a neural network to determine uplink power control parameters, utilizing as an input the selection of radio access network parameters, more than one radio access network configuration, one or more types of radio access network traffic and varying radio access network conditions;

utilizing the trained neural network, with as an input the selection of radio access network parameters, obtain as an output a set of initial uplink power control parameters;

selecting, when training the neural network, a performance metric comprising at least one of the following: user equipment normalized mean data rate, normalized cell edge rate and latency; and

selecting to the set of initial uplink power control parameters those parameters which give highest value to the selected performance metric.

7. The method of claim 6 , further comprising:

utilizing as an input in the training of the neural network at least one of the following: one or more radio access network configuration, varying radio access network conditions, network inter site distance, number of users per cell.

8. The method of claim 6 , wherein the uplink power control parameters comprise target received power P 0 , and a fractional pathloss compensation factor α.

9. A method in a radio access network, comprising:

determining, from a given set of radio access network parameters, a selection of radio access network parameters which have an effect on uplink power control;

training a neural network to determine uplink power control parameters, utilizing as an input the selection of radio access network parameters, more than one radio access network configuration, one or more types of radio access network traffic and varying radio access network conditions; and

utilizing the trained neural network, with as an input the selection of radio access network parameters, obtain as an output a set of initial uplink power control parameters,

wherein the given set of radio access network parameters comprise base station positions, estimated path loss exponent, cell-specific loads, inter site distance, average number of users per cell, network center frequency and average packet size.

10. A method for a network element in a radio access network, comprising:

receiving a set of initial uplink power control parameters provided by a trained neural network;

utilizing the set of initial uplink power control parameters in an operating radio access network and select from the received set the parameters giving best performance as the uplink power control parameters to be applied in the radio access network, wherein when utilizing the set of initial uplink power control parameters in an operating radio access network, apply each uplink power control parameter of the set a predetermined amount of time;

monitoring, when utilizing the set of initial uplink power control parameters in an operating radio access network, the performance of the network; and

rejecting, when the performance deteriorates, the current uplink power control parameters and apply utilize of the different parameters from the set of initial uplink power control parameters.

11. A computer program embodied on a non-transitory computer-readable medium, said computer program comprising instructions for causing an apparatus to perform at least the following:

determining, from a given set of radio access network parameters, a selection of radio access network parameters which have an effect on uplink power control;

training a neural network to determine uplink power control parameters, utilizing as an input the selection of radio access network parameters, more than one radio access network configuration, one or more types of radio access network traffic and varying radio access network conditions;

utilizing the trained neural network, with as an input the selection of radio access network parameters, obtain as an output a set of initial uplink power control parameters;

selecting, when training the neural network, a performance metric comprising at least one of the following: user equipment normalized mean data rate, normalized cell edge rate and latency; and

selecting to the set of initial uplink power control parameters those parameters which give highest value to the selected performance metric.

12. A computer program embodied on a non-transitory computer-readable medium, said computer program comprising instructions for causing a network element in a radio access network to perform at least the following:

receiving a set of initial uplink power control parameters provided by a trained neural network;

utilizing the set of initial uplink power control parameters in an operating radio access network and select from the received set the parameters giving best performance as the uplink power control parameters to be applied in the radio access network, wherein when utilizing the set of initial uplink power control parameters in an operating radio access network, apply each uplink power control parameter of the set a predetermined amount of time;

monitoring, when utilizing the set of initial uplink power control parameters in an operating radio access network, the performance of the network; and

rejecting, when the performance deteriorates, the current uplink power control parameters and apply utilize of the different parameters from the set of initial uplink power control parameters.