Network Configuration Based on UE Usage Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless networks face challenges in efficiently managing radio frequency resources to meet varying demands and quality of service requirements across different user equipment and networks, leading to potential congestion and degraded user experiences.
Innovation Solution
The implementation of a system that predicts the presence and usage patterns of user equipment at specific locations, generates pseudo-random usage metrics based on seed parameters, and adjusts network configuration parameters to ensure sufficient capacity and quality of service, while optimizing resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If wireless networks allocate RF resources to satisfy demand and QoS requirements, then user service quality is improved, but network congestion and resource exhaustion occur
Solution Approach 1:
The system performs preliminary actions by predicting future UE presence and usage patterns at specific locations before the actual demand occurs. This allows the network to proactively allocate RF resources in advance, ensuring QoS requirements are met while preventing congestion by distributing resources over time rather than reacting to peak demands.
Solution Approach 2:
The system creates copies of usage metrics by generating pseudo-random usage metrics based on seed parameters from observed UE behavior. These synthetic usage patterns are then aggregated to model expected network demand, allowing the system to plan resource allocation without requiring actual real-time measurements for every scenario.
2Ease of operation
If network configuration parameters are adjusted to meet user demand, then user satisfaction is improved, but network complexity increases
Solution Approach 1:
The system implements self-service by automatically predicting UE usage patterns and adjusting network configuration parameters without manual intervention. The network autonomously monitors actual usage, compares it with predicted metrics, and modifies RF resource allocation to maintain QoS, thereby improving user satisfaction while avoiding the complexity of manual network management.
Solution Approach 2:
The system establishes a feedback loop where actual UE usage metrics are continuously monitored and compared against predicted usage patterns. This feedback drives automatic adjustments to network configuration parameters, ensuring user satisfaction is maintained while the complexity of manual tuning is eliminated through automated closed-loop control.
3Productivity
If shared access to RF resources is provided for different networks, then resource utilization efficiency is improved, but network performance degradation occurs
Solution Approach 1:
The system applies segmentation by dividing RF resource allocation into separate predicted usage profiles for different networks and UE groups. By generating and aggregating pseudo-random usage metrics for each network separately, the system can allocate resources to each network in a controlled manner, maintaining both high overall utilization and individual network performance through differentiated resource management.
Data Source
AI summary
A system described herein may identify sets of seed parameters that are each associated with a respective User Equipment (“UE”) of a group of UEs. The system may generate, based on the seed parameters for each UE, a respective set of UE usage metrics for each UE, and may generate a set of aggregate UE usage metrics based on the generated sets of UE usage metrics. The system may compare the aggregate UE usage metrics to a measure of network capacity; determine an amount of time that a measure of UE usage exceeds the measure of network capacity; determine that the amount of time, that the measure of UE usage exceeds the measure of network capacity, exceeds a threshold amount of time; and modify network configuration parameters based on determining that the amount of time, that UE usage exceeds the network capacity, exceeds the threshold amount of time.