Bandwidth Allocation for Low-Latency Optical Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Passive optical networks (PONs) face challenges in meeting low-latency and high-speed requirements for 5G services due to increased latency characteristics caused by the number of optical network units (ONUs) and traffic amounts, leading to decreased network utilization and increased costs.
Innovation Solution
A bandwidth allocation method that prioritizes static bandwidth allocation for ONUs requiring low latency and dynamic bandwidth allocation for those not requiring low latency, using a service level agreement (SLA) table to manage bandwidth allocation based on queue credits and round trip times, ensuring efficient use of allocable bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static bandwidth allocation is used for all ONUs, then low-latency requirements are guaranteed, but network utilization rate decreases
Solution Approach 1:
The patent segments ONUs into two categories: low-latency required ONUs and low-latency not-required ONUs. Different bandwidth allocation methods are applied to each segment - static bandwidth allocation (SBA) for low-latency required ONUs and dynamic bandwidth allocation (DBA) for low-latency not-required ONUs. This segmentation allows the system to guarantee low latency for critical services while improving overall network utilization through efficient resource allocation to non-critical services.
2Productivity
If dynamic bandwidth allocation is used for all ONUs, then network utilization rate improves, but low-latency requirements cannot be guaranteed
Solution Approach 1:
The patent divides the ONU population into segments based on service requirements. Low-latency required ONUs receive SBA to guarantee latency performance, while low-latency not-required ONUs receive DBA to maximize network utilization. This segmentation resolves the contradiction by applying the appropriate allocation strategy to the appropriate segment.
Solution Approach 2:
The patent applies different bandwidth allocation qualities to different ONUs based on their service requirements. Low-latency required ONUs receive guaranteed static bandwidth allocation, while low-latency not-required ONUs receive dynamic allocation that adapts to traffic conditions. This local differentiation allows the system to simultaneously achieve both low latency guarantee and high network utilization.
3Quantity of substance
If the number of ONUs increases, then network capacity expands, but latency characteristic increases
Solution Approach 1:
The patent segments ONUs based on latency requirements and applies differentiated allocation strategies. By identifying which ONUs require low latency and which do not, the system can allocate resources accordingly, allowing more ONUs to be served without proportionally increasing latency for critical services.
Solution Approach 2:
The patent introduces dynamic bandwidth allocation for low-latency not-required ONUs, allowing the system to adaptively adjust bandwidth distribution based on real-time traffic conditions. This dynamic approach enables the network to accommodate more ONUs while maintaining acceptable latency for critical services through flexible resource management.
Data Source
AI summary
A bandwidth allocation apparatus and method for providing a low-latency service in an optical network that may guarantee low-latency requirements and improve a network utilization rate by allocating a static bandwidth to an ONU requiring low latency within an allocable bandwidth and by allocating a dynamic bandwidth to an ONU not requiring the low latency within a remaining bandwidth.