Bandwidth Allocation Using Prior Cycle Credits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bandwidth allocation schemes in point-to-multipoint networks are inflexible and inefficient, leading to wasteful bandwidth usage and increased queuing delays, particularly in dynamic assignment schemes that require all nodes to report their bandwidth requirements before allocation.
Innovation Solution
A flexible online bandwidth allocation scheme that distributes accumulated bandwidth credits from prior granting cycles, allowing for real-time allocation based on the needs of individual nodes, reducing the need for extensive polling and improving channel utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If dynamic bandwidth allocation schemes are used to improve network flexibility, then adaptability is improved, but device complexity increases due to extensive polling requirements
Solution Approach 1:
The patent applies preliminary action by pre-allocating bandwidth credits to ONUs based on their historical bandwidth usage patterns and service level agreements before the actual bandwidth allocation cycle. This pre-allocation eliminates the need for extensive real-time polling, as ONUs can autonomously utilize their pre-assigned credits, thereby reducing polling complexity while maintaining allocation flexibility.
Solution Approach 2:
The patent implements self-service by enabling ONUs to autonomously determine their bandwidth allocations based on pre-configured credit rules and their own queue depth information. Each ONU can independently calculate its eligible bandwidth transmission without requiring complex centralized polling and arbitration, thus reducing device complexity while preserving adaptability through rule-based dynamic allocation.
2Reliability
If traditional polling-based bandwidth assignment is used to ensure fair allocation, then reliability is improved, but loss of time increases due to extensive polling cycles
Solution Approach 1:
The patent applies preliminary action by pre-calculating and distributing bandwidth credits to ONUs based on their service level agreements and historical usage patterns before the polling cycle begins. This pre-computation eliminates the need for time-consuming real-time polling to determine allocation fairness, as fairness is already embedded in the pre-assigned credit distribution.
Solution Approach 2:
The patent implements feedback by using ONU-reported queue depth information combined with pre-assigned credit balances to dynamically adjust bandwidth allocations. This feedback mechanism ensures fair allocation by allowing ONUs with higher legitimate bandwidth needs (indicated by queue depth) to utilize more of their pre-assigned credits, while maintaining overall fairness through the credit system's inherent resource conservation properties.
3Productivity
If bandwidth is allocated based on real-time queue depth reporting to reduce queuing delays, then productivity is improved, but loss of time increases due to the reporting overhead
Solution Approach 1:
The patent implements self-service by enabling ONUs to autonomously determine their bandwidth allocations based on their own queue depth reports and pre-assigned credit balances without requiring complex centralized arbitration. Each ONU independently calculates its eligible transmission bandwidth, eliminating time-consuming polling and arbitration overhead while maintaining high productivity through efficient credit-based allocation.
Solution Approach 2:
The patent applies partial action by having ONUs report queue depth information only when necessary (e.g., when credits need replenishment or allocation adjustments are required) rather than continuously. This reduces reporting overhead and time loss while still maintaining productivity by ensuring bandwidth allocations are updated based on current network conditions when reporting occurs.
Data Source
AI summary
An apparatus comprises a receiver configured to receive a reported queue depth for a current granting cycle, and a processor coupled to the receiver and configured to determine accumulated bandwidth credits from at least one granting cycle prior to the current granting cycle. An apparatus comprises a transmitter configured to transmit a reported queue depth for a second granting cycle, a processor coupled to the transmitter, and a receiver coupled to the processor and configured to receive a transmission grant size that is based on the reported queue depth and accumulated bandwidth credits from a first granting cycle prior to the second granting cycle.