Passive Network Capacity Probing via Packet Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring network capacity, such as active probing, increase network traffic load and cannot effectively probe firewalled devices, and passive methods like TCP probing face challenges in wireless settings and long lead times for accurate capacity estimation.
Innovation Solution
A passive probing method that listens to transmitted reliable packet traffic to identify and match packet sets, calculating capacity based on registered times of suitable packets and acknowledgments, allowing for capacity measurement without generating additional probe traffic and overcoming firewalls.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If active probing methods are used to measure network capacity, then capacity measurement can be performed, but network traffic load increases and available bandwidth decreases
Solution Approach 1:
The patent applies self-service by having the network's own data packets serve as probe packets. Existing reliable packet traffic is used to perform capacity measurements without introducing external probe traffic. The packets already traversing the network for legitimate purposes are repurposed for measurement, making the system self-sufficient and eliminating the need for additional probing traffic.
Solution Approach 2:
The patent makes data packets serve multiple functions: their primary function for data transmission and an additional function for capacity measurement. By extracting timing information from legitimately transmitted packets, the system achieves dual-purpose utilization of network traffic, where the same packets provide both service data and measurement data.
2Measurement precision
If active probing methods are used to measure network capacity, then capacity measurement can be performed, but queuing times and congestion increase
Solution Approach 1:
The measurement system uses the network's existing data traffic flow to perform measurements, avoiding the need to introduce separate probe traffic that would compete for bandwidth and increase queuing times. The legitimate data packets already have priority or established flow patterns that prevent additional delays.
3Measurement precision
If active probing methods are used, then capacity measurement is possible, but firewalled devices cannot be probed
Solution Approach 1:
Firewalled devices continue to operate normally with their existing firewall rules unchanged. The measurement system passively observes packets that legitimately traverse to and from these devices, eliminating the need for devices to respond to probe requests. This approach respects firewall security policies while still enabling measurement of paths to firewalled devices.
4Adaptability or versatility
If passive TCP probing is used, then firewalled devices can be probed, but wireless settings and accurate capacity estimation face challenges
Solution Approach 1:
The patent extracts multiple timing parameters from packet flows including inter-arrival times, round-trip times, and dispersion values. By analyzing variations in these parameters across different packet sets and flow conditions, the system adapts to different network environments including wireless networks, achieving accurate capacity estimation through multi-parameter analysis rather than relying on a single measurement metric.
Data Source
AI summary
A method is described for measuring capacity on a route in a network configured to carry reliable packet traffic, by identifying and matching at least two sets of at least one first and at least one second packets transmitted on the route, the at least one first packet being transmitted in a first direction and the at least one second packet transmitted in a direction in reverse direction to the first, by also registering times for the at least two sets,and by calculating capacity of the route dependent on packet size and registered times for the at least two sets.A network node is also described. The method is advantageous for use in home networks because it uses existing traffic and therefore allows probing past device firewalls and in this instance the network node can be a gateway device, or home gateway.