Diagnosing client faults using SVM-based intelligent inference from TCP packet traces

In recent years, technological developments in computer networking have predominantly focused on improving connection media speeds and state-of-the-art applications. In tandem with user demand for high-speed delivery of information, tolerance for performance and connectivity issues has decreased. Due to the complexity and scale of modern communications networks that include a multitude of possible client devices, traditional "expert knowledge" or "rule based" methods of performance and fault diagnosis are increasingly inefficient and infeasible. Analysis of packet traces, especially from the Transmission Control Protocol (TCP), is a sophisticated inference based technique used to diagnose complicated network problems in specialized cases. TCP traces contain artifacts related to behavioral characteristics of network elements that a skilled investigator can use to infer the location and root cause of a network fault.