Security Management for Mobile Ad Hoc Network of Networks (MANoN)

Introduction

In the early part of the 21^st century, the focus of many military research efforts was on how to allow war-fighters to take advantage of all available information within the battlefield in a rapid and flexible manner. As a result, the development of the Global Information Grid (GIG) was the key enabler of this process (Stotts, Seidel, Krout, & Kolodzy, 2008). GIG is a United States (US) Department of Defense (DoD) communication project; its target is to provide agile, responsive, robust and global networking forces, sensors, users, platforms, and applications, which are used as a first step to accomplish NCW operations. NCW is a new military doctrine that seeks to translate information advantage into a competitive war-fighting advantage through the robust networking of forces distributed in large-scale conflict areas (Predd, Pfleeger, Hunker, & Bulford, 2008). In order to add to the development of the mobile networking part of the GIG, we introduced the concept of MANoN. MANoNs have various defining characteristics that differentiate them from other wired, wireless and even other ad hoc networks. MANoN is a combination of both the Mobile Ad hoc Network (MANET) (Toh, 2007) and a Network of Networks (NoN) (Spencer & Ironside, 2007; Cau, 2009), which are several nodes interconnected by wireless connections in a dynamic topology that lacks any infrastructure. Basically, each node is an ad hoc network in itself, with its own management and rules. In addition, MANoNs have the capability of operating under partial information, which makes them more flexible yet more configurable (evolvable) over time to networks joining and disconnecting, without affecting the main system. Figure 1 shows a vague idea of the GIG, consisting of different MANETs from different backgrounds and resources communicating with each other. These unique characteristics will raise non-trivial challenges for MANoNs, such as security, routing, scalability, availability, deployment considerations, media access, and Quality of Service (QoS) (Murthy & Manoj, 2004; Ilyas, 2003), in addition to conflicts which might occur because of conflicting policies (e.g. nodes following their own network policies and at the same time obeying different policies the new MANoN system might enforce) adopted by different entities in the MANoNs.

Figure 1.

Global information grid showing the mobile part MANoN

As a result, providing the components of a security management (e.g. prevention and detection) as defined by the International Telecommunication Union (2000a), is essential in order to overcome the security threats (ex. Denial of Service (DoS), host impersonation and information disclosure) our MANoN might encounter.

In this article we propose a novel, efficient, security management framework for our MANoN. We will provide a behavioral detection algorithm combined with threshold cryptography digital certificates to highlight our security administrators, and provide prevention and detection to the system. Moreover, a comprehensive, end-to-end security architecture perspective for MANoNs based on the International Telecommunication Union’s recommendations: X.800 and X.805 (1991) (2003) will be proposed.

The remainder of this article will be organized as follows: Section II will present an ITL background, and Section III will present our security management overview. Section IV will describe our scenario. Section V will explain the security architecture and its components. Section VI will illustrate and evaluate MANoN implementation using NS-2. Section VII will describe the simulation results and finally, in Section VIII we conclude the article.

Interval Temporal Logic

ITL is a linear-time temporal logic with a discrete model of time. An interval is considered to be a(n) (in)finite, non-empty sequence of states |₁ σ₂ σ₃ . . ., where a state σ_i is a mapping from the set of variables to the set of values (integers). The length |𝜎| of a finite interval 𝜎 is equal to the number of states in the interval minus one. An empty interval has exactly one state and its length is equal to 0.