Abstract:Since the local analysis is not always available, it becomes an important way to analyze the traffic pattern in remote for the network design and optimization. Apparently, it will increase the network overhead, especially for the network services nowadays which characterized with massive traffic. Hence the efficient Traffic Characteristic Data (TCD) collection is one of key technologies for the remote analysis, and then a preprocessing model for the TCD collection is presented with extracting the interesting fields from the traffic and assembling them into a Traffic Patten Package (TPP), and with an algorithm of Traffic Pattern Extracting (TPE) is designed in this paper to collect the TCDs in remote. Experimental results show that this approach has an excellent extracting performance for the remote traffic pattern analysis.