Power Distribution Settings for Multi-Radio User Equipment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in providing real-time power optimization for devices with multiple radios in advanced networks like 5G and 6G, while also managing electromagnetic exposure.
Innovation Solution
The use of reinforcement learning and predictive analysis to determine power distribution settings for user equipment with multiple radios, based on historical power usage, performance results, current location, and executing applications, while minimizing electromagnetic exposure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If power distribution settings are optimized in real-time for multiple radios, then device performance and power efficiency are improved, but system complexity and computational overhead increase
Solution Approach 1:
The system employs reinforcement learning agents that autonomously learn and optimize power distribution settings without requiring manual configuration or complex centralized control. The agents self-adjust power allocation based on real-time network conditions and device state, reducing the need for complex external management systems while maintaining optimization performance
Solution Approach 2:
The system pre-trains reinforcement learning models using historical power usage data and performance metrics before deployment. This preliminary training phase allows the system to capture complex patterns and relationships in advance, reducing the computational complexity required during real-time operation while maintaining high optimization efficiency
2Measurement precision
If historical power usage data is collected and analyzed, then predictive power optimization is improved, but data storage requirements and processing time increase
Solution Approach 1:
The system performs offline pre-training of reinforcement learning models using extensive historical power usage data. During this preliminary phase, complex patterns and relationships are captured and stored as trained model parameters. During real-time operation, the system only needs to infer predictions from the pre-trained model, dramatically reducing processing time while maintaining high predictive accuracy
Solution Approach 2:
The system creates simplified representations or copies of complex historical patterns through the reinforcement learning model. Instead of storing and processing raw historical data during operation, the system uses the trained model as a compressed representation that captures essential patterns, reducing both storage requirements and processing time while preserving predictive accuracy
3Productivity
If multiple radios operate simultaneously with optimized power distribution, then wireless service coverage and performance are improved, but electromagnetic exposure increases
Solution Approach 1:
The reinforcement learning agents optimize power distribution at the individual radio level, assigning different power levels to different radios based on specific network conditions, device state, and service requirements. This granular local optimization allows the system to minimize electromagnetic exposure from each radio individually while maintaining overall wireless service performance through coordinated operation
Solution Approach 2:
The system dynamically adjusts power distribution parameters across multiple radios based on real-time conditions. By changing power levels, frequency allocations, and modulation schemes according to network demand and environmental factors, the system maintains adequate wireless service performance while minimizing electromagnetic exposure when full power operation is not necessary
Data Source
AI summary
Facilitating real-time power optimization in advanced networks (e.g., 5G, 6G, and beyond) is provided herein. Operations of a method can include determining, by a system comprising a memory and a processor, a power distribution setting for a user equipment that includes multiple radios based on a historical radio power usage, a historical performance result, a current location, and an application currently executing on the user equipment. The method also can include implementing, by the system, the power distribution setting across the multiple radios of the user equipment. The first radio of the multiple radios can be a first radio type and a second radio of the multiple radios can be a second radio type, different from the first radio type.