Packet Flow Sampling for Overlapping Network Traffic Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The sequential approach of the Access Control List (ACL) technique for identifying packet flows in packet-switched networks fails to properly identify packets belonging to at least partially overlapping flows, requiring a large number of identification rules and additional merging steps for multilevel or multidimensional measurements.
Innovation Solution
Distribute packets into statistically uniform sample sequences and apply identification rules to each sequence, reducing the number of required rules by identifying sample sequences based on a hash function, allowing for efficient analysis of overlapping packet flows without additional merging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the sequential ACL technique is used to identify packet flows, then the identification process is simple and straightforward, but it fails to properly identify packets belonging to overlapping flows and requires a large number of identification rules
Solution Approach 1:
The patent divides the packet identification process into two independent stages: first distributing packets into sample sequences using a hash function, then applying identification rules to each sample sequence separately. This segmentation allows overlapping packet flows to be identified correctly without requiring a large number of rules, resolving the contradiction between operational simplicity and identification accuracy.
Solution Approach 2:
The patent introduces sample sequences as an intermediary structure between raw packets and final identification results. By distributing packets into statistically uniform sample sequences first, the system enables accurate identification of overlapping flows while maintaining operational simplicity, thus resolving the technical contradiction.
2Reliability
If multiple identification rules are applied to cover all packet flows including overlapping ones, then identification accuracy improves, but the number of required rules and system complexity increases significantly
Solution Approach 1:
The patent segments the identification task by first distributing packets into sample sequences via hash function, then applying identification rules to each sample sequence. This approach achieves accurate identification of overlapping packet flows with a manageable number of rules, resolving the contradiction between reliability and device complexity.
Solution Approach 2:
The patent changes the parameter of packet distribution from direct sequential processing to hash-based statistical uniform distribution. This parameter change enables accurate identification of overlapping flows while keeping the number of identification rules manageable, thus resolving the contradiction between reliability and complexity.
3Ease of operation
If the sequential ACL approach is used for multilevel or multidimensional measurements, then the processing method is straightforward, but additional merging steps are required to combine results from multiple rules
Solution Approach 1:
The patent segments the measurement process by applying identification rules to separate sample sequences, where each sequence is processed independently. This eliminates the need for additional merging steps required in sequential ACL approaches, resolving the contradiction between processing simplicity and measurement efficiency.
Solution Approach 2:
The patent performs preliminary distribution of packets into sample sequences using hash functions before applying identification rules. This preliminary action organizes the data structure to enable direct, independent processing of each sample sequence, eliminating the need for subsequent merging operations and improving measurement efficiency.
Data Source
AI summary
An apparatus and associated method for processing packets transmitted in a packet-switched communication network includes a sampling module that identifies amongst the received packets a plurality of samples distributed in a statistically uniform way amongst at least two non-overlapping sample sequences. Each sample sequence is then subjected to at least one identification rule, thereby identifying in the sample sequence at least one sub-sequence of samples fulfilling the at least one identification rule. The identification rule comprises a condition on the value of at least one identification field of the packets. Then, at least one parameter indicative of a behavior of the at least one sub-sequence of samples is provided.