Cell Reselection Parameters for 5G Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current load balancing techniques in 5G networks primarily focus on active UEs, neglecting idle mode UEs which eventually consume network bandwidth, leading to uneven traffic distribution and system performance degradation.
Innovation Solution
A neural network-based system that selects optimal cell reselection parameters using an actor-critic neural network architecture, trained through proximal policy optimization, to redistribute communication load among cells based on IP throughput and UE activity, thereby improving load balancing for idle mode UEs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If rule-based load balancing methods are used, then implementation is simple, but adaptability to traffic changes is poor
Solution Approach 1:
The patent transitions from static rule-based load balancing to dynamic reinforcement learning-based load balancing. The RL agent continuously learns from network state observations and adapts its load balancing decisions in real-time, enabling the system to dynamically respond to changing traffic patterns while maintaining computational feasibility through the actor-critic architecture.
Solution Approach 2:
The patent implements feedback mechanisms where the RL agent observes network state (including idle UE information) and receives rewards based on load balancing performance. This feedback loop enables the system to learn from past decisions and continuously improve its load balancing strategy, resolving the adaptability issue while maintaining implementation through standardized RL frameworks.
2Productivity
If existing load balancing algorithms focus on active UEs, then active traffic is optimized, but idle mode UEs are neglected leading to uneven distribution
Solution Approach 1:
The patent extends load balancing to handle both active and idle UEs universally. The RL agent observes network state that includes idle UE information and incorporates idle UE distribution into its load balancing decisions, creating a multi-functional system that optimizes for both active traffic and idle UE placement, thereby achieving more reliable overall load distribution.
Solution Approach 2:
The patent applies preliminary action by considering idle UE distribution in advance of their potential activation. By proactively balancing load based on predicted idle UE behavior rather than reacting after activation occurs, the system achieves more accurate load prediction and better overall network reliability.
3Quantity of substance
If more cells are used to service UEs, then network capacity increases, but traffic distribution becomes uneven concentrating on fewer cells
Solution Approach 1:
The patent applies local quality by making cell-specific load balancing decisions based on individual cell characteristics and current load states. The RL agent observes and responds to the specific state of each cell, assigning UEs to cells based on their individual properties rather than applying uniform treatment, thereby achieving uniform traffic distribution across all cells while maintaining high network capacity.
Data Source
AI summary
An apparatus distributing communication load over a plurality of communication cells may select action centers from random cell reselection values, based on a standard deviation of an internet protocol (IP) throughout over the plurality of communication cells; input a first vector indicating a communication state of a communication system and a second vector indicating the standard deviation of the IP throughout of the plurality of communication cells, to a neural network to output a sum of the action centers and offsets as cell reselection parameters; and transmit the cell reselection parameters to the communication system to enable a base station of the communication system to perform a cell reselection based on the cell reselection parameters.