Statistical trace-based method, apparatus, node and system for real-time traffic classification

Provided with disposed on the real-time classified to the business quantity based on offline determining business quantity classifying square device and method. Obtaining to the leads are tracking, and statistical analysis to a. The statistical analysis with the marked the business quantity parameter multi-dimensional space domain. Obtaining with marked and related classifying square, and a comprising a business quantity classifying part is used for service real-time classifying. Business quantity classifying part, is made typical of the edge network node, extracts packets in the header current of the predetermined probability. And the sampling class's statistical information in the table of tracking, streaming high-precision recorder value of the secondary provides of streaming ended current the probability of measuring duration. The tracking main high bandwidth class's table entry and square classifying end of the real-time classifying to the current sampling and state. A line, a square comprising a a movement with a matched characteristic's current flowing is. The power end; the current finished and lower overhead messages in real-time classifying to the high current bandwidth of the low overhead messages.