QoS Bandwidth Allocation With Traffic Prediction for Bursty Network Demand
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional MAC layer implementations in cable and optical data networks statically allocate bandwidth, leading to inefficiencies and unsatisfactory QoS due to inflexibility in accommodating short-term demand variations, causing packet queuing and missed QoS targets, especially under conditions of service oversubscriptions and token bucket constraints.
Innovation Solution
A QoS-based dynamic bandwidth allocation MAC layer controller with a traffic prediction module that adjusts bandwidth dynamically, incorporating a default token bucket, on-demand token replenishment, and a QoS scheduler to enforce contracted rates and adapt to short-term traffic fluctuations, ensuring statistical QoS and efficient resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If static bandwidth allocation is used in conventional MAC layer implementations, then network management is simple and contracted rates are enforced, but the system cannot accommodate short-term demand variations leading to packet queuing and missed QoS targets
Solution Approach 1:
The patent implements dynamic bandwidth allocation by continuously adjusting the allocated bandwidth based on predicted traffic patterns. The MAC layer controller monitors actual traffic and compares it with predicted values, then dynamically modifies bandwidth allocation to match actual demand while maintaining QoS guarantees. This dynamic adjustment mechanism resolves the contradiction by enabling adaptability to short-term variations without requiring complete system redesign.
Solution Approach 2:
The patent employs traffic prediction modules that forecast future traffic patterns in advance. By predicting traffic demands before they occur, the system can proactively adjust bandwidth allocation to prevent packet queuing and QoS violations. This preliminary action allows the system to accommodate short-term demand variations while maintaining manageable complexity through automated prediction algorithms.
2Productivity
If dynamic bandwidth allocation is implemented to accommodate traffic fluctuations, then QoS guarantees are maintained and resource utilization is optimized, but the system complexity increases
Solution Approach 1:
The patent implements feedback mechanisms where the MAC layer controller continuously monitors actual traffic patterns, compares them with predicted values, and adjusts bandwidth allocation accordingly. This feedback loop enables the system to optimize resource utilization by allocating bandwidth based on actual demand while maintaining QoS guarantees. The automated feedback process manages system complexity through algorithmic control rather than manual configuration.
Solution Approach 2:
The patent dynamically changes bandwidth allocation parameters based on predicted and actual traffic patterns. By adjusting allocation parameters such as bandwidth capacity and timing schedules, the system optimizes resource utilization efficiency. These parameter changes are performed automatically by the MAC layer controller, managing complexity through standardized adjustment protocols rather than complex system reconfiguration.
3Reliability
If conventional token bucket policing is used to enforce contracted rates, then rate compliance is maintained, but short-term bursts are blocked causing unnecessary packet queuing
Solution Approach 1:
The patent extends the conventional static token bucket approach by implementing dynamic token bucket policing that adjusts bucket capacity and refill rates based on predicted traffic patterns. This dynamic extension allows the system to accommodate short-term bursts within QoS guarantees by temporarily increasing bucket capacity when traffic patterns predict upcoming demand spikes, thereby reducing packet queuing delay while maintaining rate compliance reliability.
Solution Approach 2:
The patent uses traffic prediction to perform preliminary actions in preparing token buckets for upcoming traffic bursts. By predicting future traffic patterns, the system can proactively refill token buckets or adjust their capacity before actual bursts occur, preventing packet queuing and delays while maintaining enforced contracted rates. This preliminary preparation resolves the contradiction between rate compliance and burst accommodation.
Data Source
AI summary
A novel dynamic bandwidth allocation system integrates an intelligent traffic prediction module and an on-demand token replenishment module in a quality-of-service (QoS)-based media access control (MAC) layer controller, which is able to dynamically allocate data transmission capacity (i.e., bandwidth) to multiple incoming service flows of data packets for multiple subscribers more efficiently with less network congestions and service inconsistencies than conventional MAC layer management methods. The novel dynamic bandwidth allocation system utilizes network traffic predictions to identify potentially sudden or rapid changes in near-term network traffic, and if necessary, adjusts bandwidth allocations preemptively to service flows to minimize service quality degradations while also improving network resource usage efficiencies, compared to conventional static bandwidth allocation methods. The novel dynamic bandwidth allocation system also minimizes unnecessary packet droppages, network congestions, and/or speed degradations during bursty and oscillating network traffic, and improves QoS satisfaction rates for network subscribers.