5G Beam Scheduling Policy for Cross-Cell Interference Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 5G wireless networks face significant performance drops due to cross-beam inter-cell interference, which is not effectively managed by current technologies, impacting key performance indicators such as SINR and MCS.
Innovation Solution
A centralized self-organizing network entity generates cross-beam inter-cell interference profiles using machine learning models, establishing a beam scheduling policy to mitigate interference by applying penalties to co-scheduled beams, thereby optimizing beam usage across cells.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple beams are co-scheduled in the same time-frequency resources across different cells, then network capacity and resource utilization are improved, but cross-beam inter-cell interference increases causing performance degradation
Solution Approach 1:
The system implements a feedback mechanism where the centralized self-organizing network entity receives performance measurements from multiple cells, processes this information through machine learning models to generate interference profiles, and then adjusts beam scheduling policies accordingly. This closed-loop feedback system enables dynamic optimization of beam co-scheduling to maximize network capacity while maintaining acceptable interference levels.
Solution Approach 2:
A centralized self-organizing network entity acts as an intermediary between multiple base stations. This intermediary collects data from all cells, performs centralized interference analysis using machine learning, and coordinates beam scheduling across cells. By introducing this intermediary layer, the system can optimize overall network performance while managing inter-cell interference that individual cells cannot control alone.
2Device complexity
If traditional scheduling methods are used without centralized coordination, then system complexity is reduced, but interference management capability deteriorates
Solution Approach 1:
The system divides interference management functions into two segments: local scheduling decisions remain at individual base stations (maintaining simplicity), while centralized interference profile generation and policy coordination are handled by a separate self-organizing network entity. This segmentation allows complex interference management capabilities without significantly increasing the complexity of individual scheduling components.
Solution Approach 2:
The centralized self-organizing network entity serves as an intermediary that provides advanced interference management capabilities without requiring complex changes to individual base station schedulers. This intermediary handles the computationally intensive machine learning-based interference analysis, allowing local schedulers to remain relatively simple while benefiting from enhanced interference management.
3Measurement precision
If machine learning models are deployed for interference profile generation, then interference management accuracy is improved, but computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary action by continuously training and updating machine learning models using historical performance data during periods when real-time scheduling is not critical. This allows the models to be well-prepared and optimized before they are needed for real-time interference profile generation, reducing the computational burden during time-critical scheduling operations.
Solution Approach 2:
The system implements a two-level approach where machine learning models provide detailed interference profiles for critical scenarios, while simpler rules or pre-computed profiles are used for less critical cases. This partial application of complex machine learning methods reduces overall processing time while maintaining high accuracy where it matters most for network performance.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method is provided for receiving, at a data collection entity, from each cell, a time series of a respective set of data where each set of data comprises at least: one or more per-cell performance measurement data, one or more per-cell serving beams of each scheduled UE device, and a time and frequency allocation of each scheduled UE device to be served by the corresponding one or more per-cell serving beams; generating, by a cSON entity, from the sets of data received from the data collection entity, a set of cross-beam inter-cell interference profiles; establishing, by the cSON entity, from at least the set of cross-beam inter-cell interference profiles, a beam scheduling policy; receiving, at each cell, from the cSON entity, the beam scheduling policy; and applying, by a respective scheduler at each cell, the beam scheduling policy to each of the one or more per-cell serving beams.