Controlled Delay Packet Forwarding for Jitter Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional packet forwarding methods in computer networks result in significant jitter accumulation, particularly due to media access control (MAC) jitter and network queuing jitter, which exceeds the target standard deviation for Stratum 2 timing distribution, impacting the accuracy of real-time applications and services.
Innovation Solution
Implementing controlled delay queuing logic at each network switch to hold packets for a predetermined queuing time, ensuring that timing packets experience predictable and reduced jitter by applying a controlled delay greater than the maximum accumulated MAC and network queuing jitter, thereby eliminating or minimizing MAC jitter and network queuing jitter.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional store-and-forward queuing is used at each switching device, then network throughput and resource utilization are improved, but jitter accumulation increases significantly
Solution Approach 1:
The patent segments packet traffic into different classes (real-time timing packets and data packets) with different priority levels. Each class is handled by dedicated queuing logic that applies different forwarding strategies, allowing timing packets to receive predictable delay treatment while data packets maintain high throughput utilization.
Solution Approach 2:
The patent changes the delay parameter for timing packets from variable (conventional queuing) to controlled and predictable. By applying a predetermined delay value that accounts for MAC jitter and network queuing jitter, the system transforms the timing packet forwarding parameter to achieve sub-100-microsecond jitter accumulation while maintaining overall network productivity.
2Measurement precision
If GPS or Stratum 2 rubidium oscillators are deployed at each client device, then timing accuracy is improved, but system cost increases substantially
Solution Approach 1:
The patent extracts the timing accuracy requirement from individual client devices and concentrates it at the network infrastructure level. By implementing controlled delay queuing at switches and using a single Stratum 1 reference at the time server, the system removes the need for expensive timing hardware at each client while maintaining Stratum 2 quality timing distribution.
Solution Approach 2:
The patent introduces controlled delay queuing logic at network switches as an intermediary mechanism. This intermediary compensates for jitter introduced by conventional queuing, allowing timing packets to maintain accuracy without requiring expensive local oscillators at each client device.
3Measurement precision
If predetermined controlled delay is applied to timing packets, then jitter accumulation is reduced, but network latency increases
Solution Approach 1:
The patent applies different delay characteristics to different packet classes locally at each switching device. Timing packets receive predetermined controlled delay for jitter reduction, while data packets use conventional queuing for high throughput. This local differentiation allows jitter control for timing packets without significantly impacting overall network latency for data traffic.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An apparatus and method are described for forwarding of packets with controlled delay. In one embodiment, the invention includes controlled delay queuing logic to hold a packet in a first class for a queuing time of at least a controlled delay value, and scheduling logic to determine whether to delay transmission of a packet in a second class to allow the transmission of the packet in the first class when the queuing time reaches the controlled delay value.