ML Handover Parameter Estimation in Telecom Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current machine learning approaches in telecommunications networks struggle to predict optimal connection handover parameters without manufacturer-specific knowledge, leading to suboptimal performance in maintaining continuous connections during signal transitions between radio cells.
Innovation Solution
A hybrid method combining domain knowledge of 3GPP standards with artificial intelligence to estimate connection handover parameters by learning from series of signal observations, using a model that represents a function for determining these parameters, which improves prediction accuracy and flexibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a static, non-trainable function is used to determine connection handover parameters, then the system can operate without manufacturer-specific knowledge, but the prediction accuracy and adaptability to different network conditions deteriorate
Solution Approach 1:
The patent transforms static handover parameters into dynamic, learnable parameters. The machine learning model learns optimal parameter values (thresholds, hysteresis margins, timing advance values) from historical data, allowing the system to adapt to different network conditions and operator preferences without hardcoding manufacturer-specific knowledge.
Solution Approach 2:
The system performs self-learning and self-optimization by automatically training the machine learning model on collected handover data. The model continuously improves its predictions of handover parameters without requiring manual configuration or manufacturer-specific expertise, enabling the network to self-adapt to changing conditions.
2Measurement precision
If domain knowledge of 3GPP standards is combined with artificial intelligence, then the prediction accuracy of connection handover parameters improves, but the system complexity increases
Solution Approach 1:
The patent merges domain knowledge of 3GPP handover standards with artificial intelligence by integrating standard-compliant parameter structures with machine learning models. The model learns within the framework of existing standards (RSRP thresholds, hysteresis margins, timing advance) while using AI to optimize parameter selection based on actual network conditions and historical performance data.
Solution Approach 2:
The machine learning model acts as an intermediary layer between raw network measurements and handover decision parameters. It processes observed signal properties (RSRP, RSRQ, timing advance) and transforms them into optimized handover parameters, bridging the gap between measurement data and standard-compliant handover commands.
3Measurement precision
If more signal observations from multiple mobile terminals are collected, then the machine learning model accuracy improves, but the data processing requirements and system complexity increase
Solution Approach 1:
The patent segments the data collection and processing into manageable components: individual mobile terminal observations are collected separately, then aggregated for model training. The model processes observations in a structured manner, learning from patterns across multiple terminals while maintaining computational efficiency through incremental learning approaches.
Data Source
AI summary
A device and method for machine learning in a telecommunications network based on radio cells. A connection handover in the telecommunications network, in which a mobile terminal switches from one radio cell of the telecommunications network to another radio cell of the telecommunications network during a call connection or a data connection without interrupting this connection, is carried out as a function of a parameter. A series of observations of a property of a signal received by the mobile terminal in the telecommunications network is recorded. A series of observations of a signal, transmitted by a network device in the telecommunications network, for connection handover is recorded. A model for determining an estimated value for the parameter is determined as a function of the series of observations, and the estimated value is determined with the model.