Predictive Handover Management for Conflicting Cell Transfers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication networks face inefficiencies due to multiple and conflicting handovers between cells, leading to signal disruption, dropped calls, degraded data transmission, poor user experience, and battery drain, which static detection and mitigation techniques fail to address dynamically changing cell conditions and device-specific needs.
Innovation Solution
A handover management component employing machine learning models to detect, identify, and mitigate conflicting handovers by analyzing handover information, performance indicators, and dynamic conditions, predicting handover types, and determining optimal cell connections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a device is located in proximity to multiple cells, then the device can potentially connect to multiple cells for better coverage, but the device experiences multiple and conflicting handovers between cells leading to signal disruption and connection instability
Solution Approach 1:
The system performs preliminary analysis of handover patterns by examining handover information and performance indicators over a defined time period before conflicting handovers occur. This allows the system to predict potential conflicting handovers and take preventive actions, such as adjusting handover parameters or selecting optimal cells, thereby avoiding signal disruption and connection instability.
2Adaptability or versatility
If static detection techniques are used to identify conflicting handovers, then the system can detect handover patterns, but the system fails to address dynamically changing cell conditions and device-specific needs
Solution Approach 1:
The system transitions from static detection to dynamic prediction by continuously analyzing handover information and performance indicators in real-time. The machine learning model adapts to changing cell conditions and device-specific needs by learning from historical handover patterns and adjusting predictions accordingly, enabling the system to effectively manage handovers under dynamic network conditions.
3Productivity
If machine learning models are employed to predict handover types, then the system can optimize handover management, but the system complexity increases
Solution Approach 1:
The patent introduces a handover management component as an intermediary layer between the radio access network and the device. This component contains the machine learning models and handles the complex analysis of handover information and performance indicators, thereby isolating the complexity from the device itself. The intermediary processes data and generates predictions that simplify the overall system architecture while maintaining high network efficiency.
Data Source
AI summary
Enhanced management of conflicting handovers of a device between cells can be performed. Conflicting handover (CH) detector can detect conflicting handovers of the device between first cell and second cell based on handover information relating to previous handovers of device between cells, first performance indicators (PIs) associated with device and first cell, second PIs associated with device and second cell, and/or third PIs associated with second devices associated with first or second cell, over defined time period. In response to conflicting handovers detection, CH identifier can predict, from a group of types of conflicting handovers of devices, a type of the conflicting handovers of the device between first and second cells based on handover information, first PIs, second PIs, and/or third PIs. CH mitigator can determine the better of first cell or second cell to which to connect device based on rules and first, second, and/or third PIs.