ML-Based Handover Parameter Selection for Device-Specific Mobility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Handover procedures in cellular networks are resource-consuming and difficult to optimize due to momentary radio conditions, often leading to inefficient use of network resources and suboptimal user experiences.
Innovation Solution
A method using a machine-learning model in a network node to determine tailored handover parameter values for individual communication devices based on device and network context information, optimizing handover settings for specific devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional handover procedures are used with fixed parameter settings, then the handover process is simple to implement, but handover efficiency is low and network resources are wasted
Solution Approach 1:
The patent applies dynamics by transitioning from fixed handover parameters to dynamic, adaptive parameters. The machine learning model continuously learns from network data and adjusts handover parameters in real-time based on current network conditions, device characteristics, and historical performance, enabling the system to adapt optimally to changing conditions rather than relying on static configurations
Solution Approach 2:
The patent implements parameter changes by using a machine learning model to optimize handover parameters such as threshold values, hysteresis margins, and timing configurations. The model analyzes multiple parameters simultaneously and adjusts them based on learned patterns from network data, transforming the handover process from using fixed parameters to using dynamically optimized parameter sets
2Reliability
If handover parameters are optimized for each individual device, then user experience is enhanced, but the complexity of managing handover procedures increases
Solution Approach 1:
The patent applies self-service by enabling the machine learning model to autonomously optimize handover parameters without requiring manual intervention. The system automatically collects network data, trains the model, generates optimized parameter sets for individual devices, and applies these parameters autonomously, reducing the need for network operator involvement while improving handover reliability
Solution Approach 2:
The patent implements feedback by creating a closed-loop system where handover performance is continuously monitored and fed back to the machine learning model. The model uses this feedback along with network conditions and device characteristics to iteratively improve parameter optimization, creating a self-improving system that enhances reliability over time
3Loss of energy
If machine learning models are used to optimize handover parameters, then resource waste is reduced, but the computational resources required for the optimization process increase
Solution Approach 1:
The patent applies preliminary action by pre-training the machine learning model offline using historical network data before deployment. This preliminary training phase allows the model to learn optimal parameter configurations in advance, so that during actual handover operations, the model can make rapid predictions with minimal real-time computational overhead, reducing online energy consumption
Solution Approach 2:
The patent implements copying by using the machine learning model to generate optimized parameter configurations that can be replicated and applied to multiple similar devices or scenarios. Once the model learns optimal parameters for a particular device type or network condition, these optimized settings can be copied and applied broadly, reducing the need for repeated computational optimization
Data Source
Figure 1
Figure 2a
Figure 2b
AI summary
Methods and apparatuses for a network node are disclosed. According to an example, there is provided a method implemented in a network node of a communication network, the method comprising: obtaining communication device context information related to a current status of a communication device, and network context information related to a current status of the communication network; inputting the communication device context information and the network context information to a machine-learning model, wherein the machine-learning model outputs a score for at least one candidate handover parameter value based on the communication device context information and the network context information; and selecting at least one handover parameter value for a handover procedure involving the communication device based on the output from the machine-learning model, wherein the selected at least one handover parameter value is specific to the communication device.