Packet Scheduling for Real-Time Traffic in Mobile Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing packet scheduling methods in mobile telecommunication systems fail to effectively reduce packet loss rates for real-time traffic by not adequately considering future packet drop rates, leading to unsatisfied delay requirements and quality of service issues.
Innovation Solution
A method that calculates a normalized SDU drop rate based on the number of SDUs that will be dropped if not transmitted, the number of SDUs dropped until the current slot, and the maximum SDU drop rate allowed at each session, to select sessions for transmission and determine the transmission amount, reflecting future packet loss states and channel conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the scheduler selects sessions based only on current delay state and HOL packet delay, then the scheduling algorithm is simple and easy to implement, but the packet loss rate increases and delay requirements cannot be satisfied
Solution Approach 1:
The patent calculates the future SDU drop rate in advance before making scheduling decisions. By predicting how many SDUs will be dropped if not transmitted in the next slot, the scheduler proactively prevents packet loss rather than reacting to it after delays occur. This preliminary calculation of future drop rates allows the system to make informed scheduling decisions that prevent QoS violations before they happen.
Solution Approach 2:
The patent introduces dynamic adaptation by continuously updating the SDU drop rate calculation based on current buffer states, channel conditions, and historical drop patterns. The normalized SDU drop rate is dynamically adjusted for each session based on their specific QoS requirements and current traffic conditions, allowing the scheduler to adapt to changing wireless environments and traffic patterns in real-time.
2Loss of time
If the scheduler transmits packets from sessions with longer delay in buffer first, then the average delay is minimized, but the absolute delay requirement conditions cannot be satisfied
Solution Approach 1:
The patent changes the scheduling parameter from simple HOL delay to a composite metric that includes the normalized SDU drop rate. This parameter transformation allows the scheduler to consider both delay and packet loss probability together, selecting sessions based on which combination best satisfies QoS requirements rather than merely minimizing average delay.
Solution Approach 2:
The patent implements feedback by calculating the SDU drop rate based on current buffer states and channel conditions, then using this calculated drop rate to influence future scheduling decisions. The system continuously monitors transmission outcomes and adjusts scheduling priorities based on observed drop rates, creating a closed-loop control mechanism that learns from past performance.
3Productivity
If the scheduler does not reflect future packet drop rate, then the scheduling decision is fast and efficient, but the QoS requirements for real-time traffic are not satisfied
Solution Approach 1:
The patent performs preliminary calculation of the normalized SDU drop rate before the actual scheduling decision is made. By pre-calculating the expected packet loss for each session based on current buffer states and channel conditions, the system prepares scheduling metrics in advance, allowing for efficient real-time decision-making without sacrificing QoS awareness.
Solution Approach 2:
The patent calculates the SDU drop rate for all sessions in the system, even though not all sessions will be selected for transmission in the next slot. This excessive calculation ensures that when the scheduling decision is made, all necessary information is already available, preventing the need for additional calculations during time-critical scheduling moments.
Data Source
AI summary
Provided is a method of scheduling packets in a mobile telecommunication. The method includes the steps of: a) calculating a normalized standard data unit (SDU) drop rate based on the number of SDUs that will be dropped if the SDUs will not transmitted in a next slot, the number of SDUs dropped unit a current slot, the number of SDUs transmitted until the current slot and a maximum SDU drop rate allowed at each session; b) selecting sessions to transmit in a next slot and deciding a transmission amount of each session at a current slot based on the normalized SDU drop rate; and c) performing the steps a) and b) until all basic units (BU) in one slot are used.