UE Trajectory Prediction for Low-Latency Cell Handover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face challenges in efficiently managing the trajectory of user equipment (UE) in radio access networks, particularly in handling mobility and optimizing resource allocation for seamless connectivity across different cells.
Innovation Solution
Implementing artificial intelligence (AI) and machine learning (ML) techniques to predict and manage the trajectory of UE within the radio access network, utilizing AI/ML models to optimize resource allocation and handover processes based on real-time traffic patterns and device behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional wireless communication systems manage UE trajectory using conventional methods, then basic connectivity is maintained, but network efficiency is reduced and latency increases due to inability to predict mobility patterns
Solution Approach 1:
The system performs preliminary actions by predicting future UE positions and trajectories before the UE actually moves to those locations. The AI/ML models analyze historical mobility data, current position, speed, and direction to proactively determine predicted trajectories, allowing the network to prepare resource allocations and handover configurations in advance, thereby reducing latency and improving network efficiency.
2Reliability
If AI/ML models are implemented to predict UE trajectory, then network efficiency and connectivity are improved, but system complexity increases
Solution Approach 1:
The patent introduces AI/ML models as intermediary components that bridge the gap between raw UE mobility data and network decision-making systems. These models process complex trajectory predictions and translate them into actionable insights for resource allocation and handover management, thereby improving connectivity reliability while managing system complexity through specialized intermediate processing layers.
Solution Approach 2:
The system implements self-service by enabling UEs to contribute their own mobility data (position, speed, direction) that is fed into the AI/ML models. The models then use this self-provided data along with network observations to generate predictions that directly benefit the UE's own connectivity management, reducing the need for extensive external control infrastructure.
3Productivity
If real-time trajectory prediction is performed using AI/ML models, then resource allocation is optimized and handover processes are improved, but computational resources and processing power are consumed
Solution Approach 1:
The system segments the trajectory prediction and resource allocation tasks into distinct functional components. AI/ML models handle high-level trajectory prediction based on mobility patterns, while separate network entities handle specific resource allocation decisions. This segmentation allows computational work to be distributed across different processing nodes and time intervals, reducing peak computational energy consumption while maintaining allocation efficiency.
Data Source
Figure 1A~1B
Figure 2A~2B
Figure 3
AI summary
There is provided a method comprising: receiving, by a second base station from a first base station, one or more messages comprising: multiple time indications; for each time indication of the multiple time indications, a prediction of one or more candidate serving cells of a wireless device; and for each candidate serving cell of the one or more candidate serving cells, a prediction of a priority of the candidate serving cell to become a serving cell for the wireless device; and performing, by the second base station, a radio access network performance improvement decisions and/or actions using the one or more messages received from the first base station.