Method and system for user equipment pairing in full duplex networks based on machine learning

1 . A method in a base station (BS) comprising:

determining using a reinforcement learning mechanism a pair of uplink user equipment (UL UE) and downlink (DL) UE from a plurality of UEs for respectively transmitting data to the BS and receiving data from the BS at a first transmission time interval and in a first frequency channel, wherein the reinforcement learning mechanism is used to determine the pair of UL UE and DL UE based on an input regarding states of UEs served by the BS, a state of a UE within the UEs representing a level of satisfaction of the UE with a service provided by the BS, wherein the level of satisfaction is based on a Quality of service (QoS) metric associated with the service;

transmitting to the UL UE a first scheduling grant and first configuration information for transmitting data to the BS; and

transmitting to the DL UE a second scheduling grant and second configuration information for receiving data from the BS.

2 . The method of claim 1 , further comprising:

receiving data from the UL UE at the first transmission time interval and on the first frequency channel; and

transmitting data to the DL UE at the first transmission time interval and on the first frequency channel.

3 . The method of claim 2 , further comprising:

determining, based on the data received from the UL UE and the data transmitted to the DL UE, a reward for the pair of UL UE and DL UE.

4 . The method of claim 3 , further comprising:

updating the reinforcement learning mechanism based on the reward for the pair of UL UE and DL UE.

5 . The method of claim 3 , wherein the determining, based on the data received from the UL UE and the data transmitted to the DL UE, the reward for the pair of UL UE and DL UE is based on a data rate of the UL UE and an estimate of a data rate of the DL UE.

6 . The method of claim 1 , wherein the determining using the reinforcement learning mechanism the pair of UL UE and DL UE is for one of a plurality of resource blocks of the first frequency channel.

7 . The method of claim 6 , further comprising:

prior to transmitting the first and second scheduling grants and first and second configuration information determining one or more pairs of UL UE and DL UE for remaining resource blocks of the plurality of resource blocks.

8 . The method of claim 7 ,

wherein the transmitting to the UL UE the first scheduling grant and the first configuration information includes transmitting an indication of one or more resources blocks from the plurality of resource blocks that are assigned to the pair of UL UE and DL UE-pair, and

wherein the transmitting to the DL UE the second scheduling grant and second configuration information includes transmitting the indication of the one or more resources blocks from the plurality of resource blocks that are assigned to the pair of UL UE and DL UE.

9 . A non-transitory machine-readable medium comprising computer program code which when executed by a computer, is capable of performing:

determining using a reinforcement learning mechanism a pair of uplink user equipment (UL UE) and downlink (DL) UE from a plurality of UEs for respectively transmitting data to a base station (BS) and receiving data from the BS at a first transmission time interval and in a first frequency channel, wherein the reinforcement learning mechanism is used to determine the pair of UL UE and DL UE based on an input regarding states of UEs served by the BS, a state of a UE within the UEs representing a level of satisfaction of the UE with a service provided by the BS, wherein the level of satisfaction is based on a Quality of service (QoS) metric associated with the service;

transmitting to the UL UE a first scheduling grant and first configuration information for transmitting data to the BS; and

transmitting to the DL UE a second scheduling grant and second configuration information for receiving data from the BS.

10 . A base station (BS) comprising:

a non-transitory machine-readable storage medium that provides instructions that, if executed by a processor, will cause the BS to perform operations comprising,

determining using a reinforcement learning mechanism a pair of uplink user equipment (UL UE) and downlink (DL) UE from a plurality of UEs for respectively transmitting data to the BS and receiving data from the BS at a first transmission time interval and in a first frequency channel, wherein the reinforcement learning mechanism is used to determine the pair of UL UE and DL UE based on an input regarding states of UEs served by the BS, a state of a UE within the UEs representing a level of satisfaction of the UE with a service provided by the BS, wherein the level of satisfaction is based on a Quality of service (QoS) metric associated with the service,

transmitting to the UL UE a first scheduling grant and first configuration information for transmitting data to the BS, and

transmitting to the DL UE a second scheduling grant and second configuration information for receiving data from the BS.

11 . The base station of claim 10 , wherein the operations further comprise:

receiving data from the UL UE at the first transmission time interval and on the first frequency channel; and

transmitting data to the DL UE at the first transmission time interval and on the first frequency channel.

12 . The base station of claim 11 , wherein the operations further comprise:

determining, based on the data received from the UL UE and the data transmitted to the DL UE, a reward for the pair of UL UE and DL UE.

13 . The base station of claim 12 , wherein the operations further comprise:

updating the reinforcement learning mechanism based on the reward for the pair of UL UE and DL UE.

14 . The base station of claim 12 , wherein the determining, based on the data received from the UL UE and the data transmitted to the DL UE, a reward for the pair of UL UE and DL UE is based on a data rate of the UL UE and an estimate of a data rate of the DL UE.

15 . The base station of claim 10 , wherein the determining using the reinforcement learning mechanism the pair of UL UE and DL UE is for one of a plurality of resource blocks of the first frequency channel.

16 . The base station of claim 15 , wherein the operations further comprise prior to transmitting the first and second scheduling grants and first and second configuration information:

determining one or more pairs of UL UE and DL UE for remaining resource blocks of the plurality of resource blocks.

17 . The base station of claim 16 ,

wherein the transmitting to the UL UE the first scheduling grant and the first configuration information includes transmitting an indication of one or more resources blocks from the plurality of resource blocks that are assigned to the UL UE and DL UE pair; and

wherein the transmitting to the DL UE the second scheduling grant and second configuration information includes transmitting the indication of the one or more resources blocks from the plurality of resource blocks that are assigned to the pair of UL UE and DL UE.

18 . The non-transitory machine-readable medium of claim 9 , wherein the computer program code, when executed by the computer, is capable of further performing:

receiving data from the UL UE at the first transmission time interval and on the first frequency channel; and

transmitting data to the DL UE at the first transmission time interval and on the first frequency channel.