PRIVACY-PRESERVING AUTONOMOUS SYSTEM ROUTING VIA INTELLIGENT GRAPH FILTERING

Abstract

Traditional inter-domain routing protocols' decentralized architecture can cause a number of problems, such as misconfiguration and convergence problems. Alternative strategies that use the Software Defined Networking (SDN) paradigm to give more control over routing operations have been put forth recently in response to these issues. In this scenario, an SDN controller is assigned to handle routing duties in a multidomain network made up of Autonomous Systems (ASs). Each controller must learn how to connect to any node outside of its domain in order to carry out inter-domain routing. Because the controllers must access sensitive, business-critical data (such link charges) across all domains, serious privacy issues arise. In order to preserve privacy, protocols for determining the shortest path between a source and a destination a typical policy in routing tasks have recently been presented. These protocols rely on MultiParty Computation (MPC) techniques, which limit scalability by ensuring anonymity at the expense of high computational and communication complexity. In this study, we use Data Mining (DM) approaches to eliminate nodes that have a low probability of being reached by the shortest path, thus pruning the network graph. On the pruned graph, privacy-preserving shortest path methods are then conducted at a significantly reduced complexity.