Dynamic Latency Traffic Flow Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current traffic engineering methods in packet-based networks, such as MPLS, VoIP, and IMS, fail to accurately account for variable queuing delays, leading to inaccurate latency estimation and suboptimal network design, especially for latency-sensitive services like VoIP, as they only consider static delays from non-packet based equipment.
Innovation Solution
The method employs dynamic latency, which combines static delay from non-packet based equipment with queuing delay from packet based equipment, extending the shortest path algorithm to account for varying network loads, thereby optimizing network traffic engineering and design by ensuring adequate capacity for latency-sensitive services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the shortest path method using static latency is used, then the network design is simple and computationally efficient, but the latency estimation is inaccurate for packet-based networks where queuing delays vary with network loading
Solution Approach 1:
The patent applies dynamics by transitioning from static latency values to dynamic latency values that change with network loading conditions. The shortest path algorithm is modified to use latency values that are updated based on current network traffic patterns, allowing the system to adapt to varying network conditions while maintaining the core shortest path optimization framework.
Solution Approach 2:
The patent changes the key parameter from static latency to dynamic latency. By introducing latency as a variable parameter that depends on network loading, the system achieves more accurate latency estimation for packet-based networks. The algorithm recalculates path costs based on current latency measurements, enabling accurate QoS planning without requiring complete redesign of the routing framework.
2Reliability
If dynamic latency is used to account for queuing delays, then the latency estimation accuracy improves, but the computational complexity and data processing requirements increase
Solution Approach 1:
The patent implements feedback mechanisms where latency measurements from the network are continuously collected and fed back into the latency planning algorithm. This feedback loop allows the system to update its understanding of network conditions and adjust latency allocations accordingly, ensuring reliable QoS guarantees while automating the complexity of dynamic parameter management.
Solution Approach 2:
The patent performs preliminary latency planning by pre-calculating latency budgets for different traffic flows based on historical or predicted network conditions. This preliminary action allows the system to establish QoS guarantees in advance, reducing the need for complex real-time adjustments while maintaining reliable service quality.
Data Source
AI summary
A method and apparatus that use dynamic latency to estimate traffic flows in a communication network are disclosed. For example, the method obtains a demand matrix having at least one point pair demand, and satisfies each of the at least one point pair demand in the demand matrix by selecting a shortest path that minimizes a dynamic latency and a number of hops.