Traffic Modeling Project

This is a traffic modeling engine suite of software tools. The input is a captured packet trace from a given point in the network. Models are application agnostic in that they are inferred from the connection sessions in the pcap file.


The project has been supported in part by the National Science Foundation award no. 1908974 CISE-CNS grant award.

The traffic modeling engine is developed in python and uses scipy and scikit-learn machine learning libraries.


Converts raw pcap traces into a CSV file, with all decoded headers.


Extracts a dataset from a raw packet trace file that are used by the traffic modeling engine


Generates traffic models using the datasets (or directly from a packet trace file).


Simulates the generation of traffic protocol data units (PDUs) on a
network using a traffic model.


Generates realistic network traffic on a device by using a traffic model.


Provides analysis output for multiple network trace files, comparing them based on the
metrics calculated from each input trace.

Traffic Generator Survey

Network traffic workloads are widely utilized in applied research to verify correctness and to measure the impact of novel algorithms, protocols, and network functions. We conducted a comprehensive survey of traffic generators referenced by researchers over the last 13 years, providing in-depth classification of the functional behaviors of the most frequently cited generators.

The survey paper is published at the ACM Computing Surveys.

We developed a tool to analyze about 7000 papers to identify popular traffic generators utilized in research.