Wireless Scheduling Algorithm for Fairness and Throughput
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In wireless communication systems, the proportional fairness algorithm can lead to unfairness where high data rate devices are starved of data due to excessive allocation of timeslots to low data rate devices, resulting in slow or jittery data receipt for high data rate devices, especially in areas with many low data rate devices.
Innovation Solution
The method involves detecting when a small subset of devices is monopolizing timeslots and de-scheduling low data rate devices to newly schedule high data rate devices, thereby increasing the number of devices served in a series of timeslots, preventing monopolization and ensuring fair data distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the proportional fairness algorithm allocates timeslots based on average throughput, then low data rate devices receive more scheduling priority, but high data rate devices experience prolonged waiting times and data starvation
Solution Approach 1:
The patent changes the scheduling parameter from average throughput alone to a composite metric that incorporates both average throughput and forward link data rate. This parameter modification allows the system to differentiate between devices with poor RF conditions (low data rate) and those with good RF conditions (high data rate), preventing high data rate devices from being starved while still providing assistance to low data rate devices.
Solution Approach 2:
The scheduling algorithm dynamically adjusts the weighting factors (alpha and beta) based on system conditions and device characteristics. This dynamic adjustment allows the system to flexibly balance between fairness for low data rate devices and efficiency for high data rate devices, preventing any single device type from monopolizing or being starved of timeslots.
2Reliability
If many timeslots are allocated to low data rate devices for error correction transmissions, then successful data delivery to low data rate devices is achieved, but high data rate devices are starved of data transmission opportunities
Solution Approach 1:
The patent introduces forward link data rate as an additional scheduling parameter alongside average throughput. This allows the system to recognize that high data rate devices can successfully receive data in fewer timeslots due to better RF conditions and lower error correction requirements, thereby preventing timeslot monopolization by low data rate devices while maintaining reliable delivery for them.
Solution Approach 2:
The scheduling algorithm applies partial assistance to low data rate devices by allocating them additional timeslots when needed, rather than guaranteeing excessive timeslots that would starve high data rate devices. The system provides just enough resources to low data rate devices to achieve successful transmission without compromising overall system productivity.
3Productivity
If the RAN schedules transmissions to maximize utilization of upcoming timeslots, then overall system efficiency is improved, but individual devices may experience unfair waiting times
Solution Approach 1:
The patent modifies the scheduling decision parameters to include both average throughput and forward link data rate, creating a more nuanced scheduling metric. This allows the RAN to make scheduling decisions that simultaneously optimize for system-wide efficiency and individual device fairness, avoiding the extremes of pure efficiency maximization or simple proportional fairness.
Solution Approach 2:
The scheduling algorithm incorporates feedback from historical transmission performance (average throughput) and current channel conditions (forward link data rate) to make informed scheduling decisions. This feedback mechanism enables the system to adapt to changing conditions and maintain both efficiency and fairness over time.
Data Source
AI summary
A method and system is disclosed for managing transmission of data in upcoming timeslots on an air interface. A radio access network detects that in the upcoming timeslots, less than a threshold portion of currently-contending devices are scheduled to receive transmissions, and the radio access network responsively de-schedules a number of the devices and schedules in their place a greater number of other devices, where the newly scheduled devices have higher forward link data rates than the de-scheduled devices.