ML-Driven Handover Priority Updates in 5G Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual prioritization of neighboring cells in 5G networks for handover is cumbersome and fails to adapt to changing cell properties, leading to increased handover failures.
Innovation Solution
A computer-implemented method using a machine learning engine to continuously analyze handover failure data, compute probabilities for successful handover, and automatically update the Neighbor Cell Relation Table with priorities, prioritizing cells with higher success rates for handover.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual prioritization of neighboring cells is used in the Neighbor Cell Relation Table, then handover control can be performed, but the process becomes cumbersome and fails to adapt to changing cell properties, leading to increased handover failures
Solution Approach 1:
The system performs self-optimization by automatically computing handover priorities based on historical handover data and machine learning algorithms. The gNodeB autonomously updates the Neighbor Cell Relation Table without manual intervention, allowing the network to adapt to changing conditions while reducing operational burden.
Solution Approach 2:
The system continuously collects handover data from the network and uses machine learning models to compute updated priorities. This feedback loop enables the system to learn from past handover outcomes and dynamically adjust priorities to improve success rates while adapting to changing cell properties.
2Adaptability or versatility
If manual prioritization of neighboring cells is used, then initial handover control is possible, but the priorities cannot adapt to dynamic changes in cell properties, leading to handover failures
Solution Approach 1:
The handover priorities are transformed from static manual values to dynamic computed values that automatically adapt to changing cell properties. The machine learning model continuously updates priorities based on current network conditions, historical data, and learned patterns, enabling the system to respond to dynamic changes without manual intervention.
Solution Approach 2:
The manual mechanical process of updating priorities is replaced with an automated computational system using machine learning algorithms. This substitution enables automatic adaptation to changing cell properties by processing handover data and computing optimal priorities through mathematical models rather than human operators.
3Reliability
If no prioritizing function is implemented, then the Neighbor Cell Relation Table can be automatically filled by ANR feature, but handover failures increase due to lack of intelligent selection
Solution Approach 1:
The system performs preliminary computation of handover priorities using machine learning models before actual handover decisions are made. By pre-processing historical handover data and computing probability-based priorities in advance, the system prepares optimized handover recommendations that improve success rates when handovers are needed.
Solution Approach 2:
The system changes the parameter representation from simple manual priority values to computed probability values derived from machine learning models. These probability parameters reflect the actual likelihood of successful handover based on historical data, enabling more intelligent and reliable handover decisions compared to static priority values.
Data Source
AI summary
Computer-implemented method for controlling the handover of at least one User Equipment on a moving entity, such as on a vehicle (AGV), from one cell of a mobile communication system according to fifth-generation technology standard to a neighboring cell of the mobile communication system, using a Neighbor Cell Relation Table (NCRT) which lists possible neighboring cells, comprising the steps of - continuously obtaining handover failure data of a multitude of cells of the mobile communication system, - continuously employing a machine learning engine (MLE) that computes a probability for successful handover from at least one cell of the mobile communication system to each of at least two neighboring cells of the mobile communication system, based on the obtained handover failure data, - continuously updating the Neighbor Cell Relation Table (NCRT) with priorities (PRIO) for the at least two neighboring cells of the mobile communication system which priorities are deduced from the probability for successful handover and which priorities are higher for higher probability for successful handover and lower for lower probability for successful handover, - in case of handover from one cell to one of the neighboring cells choosing the neighboring cell with the highest probability for successful handover. By automatically generating and updating the priorities for neighboring cells handover failures can be reduced.