18033407. CENTRAL NODE AND A METHOD FOR REINFORCEMENT LEARNING IN A RADIO ACCESS NETWORK simplified abstract (Telefonaktiebolaget LM Ericsson (publ))

From WikiPatents
Jump to navigation Jump to search

CENTRAL NODE AND A METHOD FOR REINFORCEMENT LEARNING IN A RADIO ACCESS NETWORK

Organization Name

Telefonaktiebolaget LM Ericsson (publ)

Inventor(s)

Yu Wang of Solna (SE)

Wenfeng Hu of Täby (SE)

Vidit Saxena of Järfälla (SE)

Pablo Soldati of Solna (SE)

CENTRAL NODE AND A METHOD FOR REINFORCEMENT LEARNING IN A RADIO ACCESS NETWORK - A simplified explanation of the abstract

This abstract first appeared for US patent application 18033407 titled 'CENTRAL NODE AND A METHOD FOR REINFORCEMENT LEARNING IN A RADIO ACCESS NETWORK

Simplified Explanation

The abstract describes a method for controlling an exploration strategy in a distributed node in a Radio Access Network (RAN) using Reinforcement Learning (RL). The central node evaluates the cost and performance of actions performed for exploration in RL modules and determines exploration parameters based on the evaluation. The central node then configures the RL modules with the determined exploration parameters to update their exploration strategy and produce data samples.

  • The method controls an exploration strategy in RL modules in a distributed node in a RAN.
  • The central node evaluates the cost and performance of actions performed for exploration in the RL modules.
  • Based on the evaluation, the central node determines exploration parameters associated with the exploration strategy.
  • The central node configures the RL modules with the determined exploration parameters to update their exploration strategy.
  • The updated exploration strategy enforces the RL modules to act accordingly and produce data samples.

Potential Applications

  • This method can be applied in various wireless communication systems that utilize RL for optimization and decision-making.
  • It can be used in network management and resource allocation in RANs to improve performance and efficiency.

Problems Solved

  • The method solves the problem of controlling and optimizing the exploration strategy in RL modules in a distributed node.
  • It addresses the challenge of evaluating the cost and performance of exploration actions and determining suitable exploration parameters.

Benefits

  • The method allows for efficient control and optimization of the exploration strategy in RL modules.
  • It enables improved performance and decision-making in wireless communication systems.
  • The method can lead to enhanced resource allocation and network management in RANs.


Original Abstract Submitted

A method performed by a central node for controlling an exploration strategy associated to Reinforcement Learning, RL, in one or more RL modules in a distributed node in a Radio Access Network, RAN, is provided. The central node evaluates a cost of actions performed for explorations in the one or more RL modules, and a performance of the one or more RL modules. Based on the evaluation, the central node determines one or more exploration parameters associated to the exploration strategy. The central node controls the exploration strategy by configuring the one or more RL modules with the determined one or more exploration parameters to update its exploration strategy, enforcing the respective one or more RL modules to act according to the updated exploration strategy to produce data samples for the one or more RL modules in the distributed node.