Machine-Learning Carrier Aggregation for Voice Quality Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Carrier aggregation in communication networks can adversely impact voice service performance due to increased interference, limited uplink power, suboptimal resource allocation, and network congestion, particularly in scenarios where data and voice services are simultaneously used.
Innovation Solution
A supervised machine learning-based classification model is employed to dynamically manage carrier aggregation by predicting voice quality, allowing or disallowing its use based on real-time network and user equipment key performance indicators, ensuring voice quality meets predefined thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If carrier aggregation is used to enhance data rates and network capacity, then data service performance is improved, but voice service performance deteriorates due to increased interference and limited uplink power
Solution Approach 1:
The patent implements dynamic carrier aggregation management where the system continuously monitors voice quality metrics and network conditions, then adjusts carrier aggregation configuration in real-time. When voice quality degradation is detected, the system dynamically releases secondary cells or adjusts uplink power allocation, transforming the static carrier aggregation setup into an adaptive system that responds to changing conditions.
Solution Approach 2:
The system changes key parameters including uplink power allocation between primary and secondary cells, carrier aggregation configuration (adding or releasing secondary cells), and resource allocation ratios. These parameter adjustments are made based on monitored voice quality metrics to optimize the trade-off between data rate and voice quality.
2Productivity
If uplink power is allocated to support carrier aggregation for data services, then network capacity is improved, but voice quality deteriorates due to power limitations
Solution Approach 1:
The system dynamically adjusts uplink power allocation parameters between primary and secondary cells based on voice quality requirements. When voice quality degradation is detected, the system increases power allocation to the primary cell and reduces or eliminates power to secondary cells, thereby maintaining voice quality while managing network capacity.
Solution Approach 2:
The patent implements a feedback mechanism where voice quality metrics are continuously monitored and fed back to the network controller. This feedback loop enables the system to adjust power allocation and carrier aggregation configuration in response to actual voice quality conditions, ensuring that power limitations do not degrade voice service.
3Productivity
If secondary cells are configured for carrier aggregation, then data service throughput is improved, but resource allocation becomes suboptimal for voice services
Solution Approach 1:
The system transforms static resource allocation into a dynamic process by continuously monitoring voice quality and network conditions. Secondary cells are added or released based on real-time conditions, and resource allocation ratios between primary and secondary cells are adjusted dynamically to optimize both data throughput and voice service quality.
Solution Approach 2:
The patent applies different resource allocation strategies to different cells based on their specific roles and conditions. The primary cell receives prioritized resources for voice services when needed, while secondary cells focus on data services. This localized quality adjustment ensures optimal resource allocation efficiency for each cell's specific function.
Data Source
AI summary
The technology described herein is directed towards a carrier aggregation system for a user equipment requesting simultaneous voice and data sessions. In one example, a classification model determines whether carrier aggregation, if used in conjunction with voice quality data, will likely result in acceptable or unacceptable voice quality for a user equipment, based on dynamic user equipment-related and cell-related input data. If voice quality is deemed acceptable by the model, carrier aggregation is allowed to be used simultaneously (and activated if not in use) with voice service. If voice quality is deemed unacceptable, carrier aggregation is not allowed to be used simultaneously with voice service, (and released if currently in use). Monitoring is performed to evaluate whether subsequent conditions (updated input data) change, such as if estimated voice quality degrades such that carrier aggregation is no longer allowed to continue when both carrier aggregation and voice are otherwise in use.