Abstract: The technology described herein is directed towards dynamically determining, based on current environment state data, a number of coverage zones within a base station's coverage area, and physical uplink shared channel (PUSCH) closed-loop power control-related data for user equipment in each zone. In one implementation, a deep reinforcement learning (DRL)-based system includes a first DRL agent that, based on the current environment state, outputs the optimal number of zones. Based on the current environment state data and the number of zones, a second DRL agent outputs optimal per-zone target signal-to-interference-plus-noise ratio (SINR) values. The SINR values are used in outputting transmit power control data to the UEs in each coverage zone. Also described is deep reinforcement learning by the system based on a reward function that can balance enhanced power efficiency with enhanced throughput to determine the optimal number of zones and SINR values for various current environment state data.

Abstract: Distributed unit monitoring of user equipment transmitter power degradation (e.g., using a computerized tool), is enabled. For example, a system can comprise at least one processor and at least one memory that stores executable instructions that, when executed by the processor, facilitate performance of operations.

Abstract: A system can communicate broadband cellular communications with a group of user equipment, wherein respective user equipment of the group of user equipment execute respective distributed deep reinforcement learning agents that are configured to determine respective transmission power levels for respective physical uplink shared channel transmissions based on respective local observations and respective learned policies. The system can determine, using a centralized deep reinforcement learning agent, respective transmission power control commands for the respective user equipment, the respective user equipment utilizing the respective transmission power control commands as constraints in the respective distributed deep reinforcement learning agents. The system can receive the respective physical uplink shared channel transmissions from the respective user equipment according to the respective transmission power levels.

Abstract: A system can, for respective user equipment (UE) of a group of UE for which broadband cellular communications are being facilitated via a network comprising the system, determine respective establishment causes, determine respective bearer types, and determine respective priority levels, wherein a first weight associated with the respective establishment causes is greater than a second weight associated with the respective bearer types, which is greater than a third weight associated with a priority level. The system can rank the respective UE based on the respective establishment causes, the respective bearer types, the respective priority levels, the first weight, the second weight, and the third weight to produce respective rankings of the respective user equipment. The system can, in response to determining that a serving-capacity criterion is satisfied with respect to the group of UE and based on the respective rankings, prioritize service offered to the respective UE.

Abstract: Automatic optimization of the nominal power parameter in uplink power control (e.g., using a computerized tool), is enabled. For example, a system can comprise a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations. The operations can comprise determining ratio data representative of a received signal to interference and noise ratio of a transmission, of a group of cellular transmissions, via a cellular node. The operations can further comprise, based on the ratio data, determining a power control metric applicable to uplink power control for user equipment. The operations can further comprise, based on the power control metric, determining a nominal power parameter for each uplink channel involving the cellular node. The operations can further comprise applying the nominal power parameter to subsequent connections made via the cellular node subsequent to the determining of the nominal power parameter.

Abstract: The technology described herein is directed towards dynamically determining, based on current environment state data, a number of coverage zones within a base station's coverage area, and physical uplink shared channel (PUSCH) closed-loop power control-related data for user equipment in each zone. In one implementation, a deep reinforcement learning (DRL)-based system includes a first DRL agent that, based on the current environment state, outputs the optimal number of zones. Based on the current environment state data and the number of zones, a second DRL agent outputs optimal per-zone target signal-to-interference-plus-noise ratio (SINR) values. The SINR values are used in outputting transmit power control data to the UEs in each coverage zone. Also described is deep reinforcement learning by the system based on a reward function that can balance enhanced power efficiency with enhanced throughput to determine the optimal number of zones and SINR values for various current environment state data.