Multi-RAT DSS Policy Using UE Scaling for Spectral Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dynamic spectrum sharing (DSS) policies between multiple radio access technologies (RATs) are inefficient, leading to lower spectral efficiency, degradation of user experience, and persistent transmission/reception errors due to reliance on pre-defined resource sharing patterns that do not account for dynamic traffic conditions and UE requirements.
Innovation Solution
Implement a reinforcement learning (RL) based approach to determine a DSS policy by calculating UE resource scaling factors, cumulative radio resource requirements, and DSS scaling factors, using a centralized node to optimize radio resource allocation across multiple RATs, incorporating UE and cell KPI metrics to enhance spectral efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If pre-defined resource sharing patterns are used for DSS policy, then implementation complexity is reduced, but spectral efficiency deteriorates
Solution Approach 1:
The patent applies dynamics by replacing static pre-defined resource sharing patterns with a dynamic RL-based framework that continuously adapts DSS policies according to real-time traffic conditions, UE requirements, and channel state information. The RL agent learns from historical data and adjusts resource allocation dynamically, transforming the system from rigid to flexible while improving spectral efficiency.
Solution Approach 2:
The patent implements feedback mechanisms where the RL agent receives feedback from KPI metrics (throughput, latency, error rates) and historical performance data. This feedback loop enables the system to learn from past decisions, adjust future resource allocation strategies, and continuously optimize spectral efficiency based on actual system performance rather than relying on fixed patterns.
2Ease of operation
If pre-defined resource sharing patterns are used for DSS policy, then system simplicity is maintained, but user experience deteriorates
Solution Approach 1:
The patent applies self-service by enabling the RL system to autonomously manage resource allocation without requiring manual configuration or complex user input. The system automatically learns user requirements and traffic patterns, making decisions about resource sharing based on real-time conditions. This autonomous operation maintains simplicity for users while significantly improving reliability and user experience through adaptive optimization.
3Device complexity
If fixed linear dimensions of traffic requirement are used, then calculation complexity is reduced, but transmission errors increase
Solution Approach 1:
The patent applies parameter changes by transitioning from fixed linear traffic requirement models to dynamic, multi-dimensional parameter representations. The system uses RL to learn and adapt traffic characteristics in real-time, adjusting allocation decisions based on actual channel conditions and user needs. This dynamic parameter adjustment increases calculation complexity but dramatically reduces transmission errors by aligning resource allocation with real-time traffic demands.
Data Source
AI summary
A method may include determining a dynamic spectrum sharing (DSS) policy in a two stage approach. In a stage, the method may include determining radio resource requirements of each user equipment (UE) among active UEs using an individual UE resource scaling factor. In another stage, the method may include determining the DSS policy based on cumulative radio resource requirements of the active UEs determined based on the individual UE resource scaling factor, and DSS scaling factors.