Wireless Resource Scheduler Optimizing Throughput and Fairness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless communication systems face challenges in achieving an optimal trade-off between efficiency and fairness in resource allocation, as maximizing efficiency often favors closer users at the expense of fairness, while maximizing fairness results in low system efficiency.
Innovation Solution
The method involves determining Jain's fairness index (JFI) and maximizing the sum of throughputs by selecting a suitable tuning parameter in an efficiency and fairness trade-off relation model, allocating resources to achieve an optimal balance between efficiency and fairness for multiple users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If resources are allocated to maximize sum of rates (efficiency), then system efficiency is improved, but fairness among users deteriorates
Solution Approach 1:
The patent introduces a tuning parameter β that changes the fairness sensitivity in the utility function. By adjusting β, the system can move along the efficiency-fairness trade-off curve to achieve different operating points, resolving the contradiction between maximizing efficiency and maintaining fairness.
Solution Approach 2:
The patent employs dynamic resource allocation where the scheduler continuously adjusts resource distribution based on real-time channel conditions and user requirements. This dynamic approach allows the system to adapt to changing conditions and achieve optimal efficiency-fairness trade-offs that are not possible with static allocation schemes.
2Reliability
If resources are allocated to maximize minimum rate (fairness), then fairness is improved, but system efficiency deteriorates
Solution Approach 1:
The patent uses the tuning parameter β to control the degree of fairness enforcement. When β is set to emphasize fairness, the system prioritizes minimum rate guarantees while still allowing efficient resource utilization through the structured utility maximization approach.
Solution Approach 2:
The patent segments the user base into different priority classes or groups based on their fairness requirements and channel conditions. This segmentation allows the system to apply different allocation strategies to different user groups, simultaneously achieving fairness for protected users and efficiency for others.
3Reliability
If a trade-off policy is used to balance efficiency and fairness, then both goals are partially achieved, but resource allocation becomes wasteful and suboptimal
Solution Approach 1:
The patent implements feedback mechanisms where the scheduler continuously monitors actual throughput and fairness metrics, comparing them against target values. This feedback loop enables real-time adjustment of resource allocation to eliminate waste and achieve optimal efficiency-fairness trade-offs that adapt to actual system conditions.
Solution Approach 2:
The patent performs preliminary characterization of the efficiency-fairness trade-off curve through channel measurements and simulations before actual resource allocation occurs. This pre-characterization allows the system to pre-compute optimal β values and allocation strategies, avoiding wasteful trial-and-error during operation.
Data Source
AI summary
Embodiments are provided for scheduling resources considering data rate-efficiency and fairness trade-off. A value of Jain's fairness index (JFI) is determined for transmitting a service to a plurality of users, and accordingly a sum of throughputs is maximized for transmitting the service to the users. Alternatively, a sum of throughputs is determined first and accordingly the JFI is maximized. Maximizing the sum of throughputs or JFI includes selecting a suitable value for a tuning parameter in an efficiency and fairness trade-off relation model. In accordance with the values of sum of throughputs and JFI, a plurality of resources are allocated for transmitting the service to the users. For static or quasi-static channels, the relation model is a convex function with a monotonic trade-off property. For ergodic time varying channels, the tuning parameter is selected by solving the relation model using a gradient-based approach.