End-to-end flow inference of encrypted MANET

Abstract

Existing traffic analysis techniques are limited by the application environment and data rate when the end-to-end encryption and link encryption are both employed in mobile Ad Hoc networks (MANETs). An end-to-end traffic analysis algorithm is proposed based on time sequence matching and correlation analysis in this paper. According to the characteristics of data flow, the algorithm adopts time sequence matching, correlation analysis, and flow tracer to find the end-to-end flow in the network. The method can infer the traffic pattern without decrypting the frames. Simulation results proved the effectiveness of the method.