Policy management in airborne tactical networks

Abstract

Policy management in a traditional IP network is accomplished by employing Policy Enforcement Points (PEP) and Policy Decision Points (PDP). Policy management has a configuration model and an outsourcing model. In a configuration model, a PDP will send new policies to a PEP to enforce. In an outsourcing model, a PEP will contact a PDP for a decision as to admit or deny traffic. In dynamic airborne tactical edge networks, fast-moving aircraft and terrain obscuration combine to create conditions where traditional policy management may be less effective to be deployed and fielded. In this paper, we study the efficacy of the configuration model in a dynamic Mobile Ad-hoc Network (MANET) by trying to answer the following questions: How quickly can we rectify a network problem by deploying new policies? Is it possible to deploy a new policy if the problem restricts communication between PDP and PEP? We will also study a case where a PEP will contact a PDP to decide on granting or denying a resource reservation request as applicable to the Resource Reservation Protocol (RSVP). This study will also consider the usage of advanced and aggregated RSVP (A-RSVP and AR-RSVP) protocols for making resource reservations. Modifications to the policy enforcement algorithms will be proposed to improve and meet end-to-end Service Level Agreements. The performance analysis in a simulation environment will be conducted for various tactical edge network scenarios.