Modeling and analyzing mobile ad hoc networks in Real-Time Maude

Abstract

Modeling and analyzing mobile ad hoc networks (MANETs) pose non-trivial challenges to formal methods. Time, geometry, communication delays and failures, mobility, and uni-and bidirectional wireless communication can interact in unforeseen ways that are hard to model and analyze by current process calculi and automatic formal methods. As a consequence, current analyses tend to abstract away these physical aspects, so thatalthough still quite useful in finding various errors-their simplifying assumptions can easily fail to model details of MANET behavior relevant to meet desired requirements. In this work we present a formal framework for the modeling and analysis of MANETS based on Real-Time Maude to address this challenge. Specifically, we show that our framework has good expressive power to model relevant aspects of MANETs, and good compositionality properties, so that a MANET protocol can be easily composed with various models of mobility and with other MANET protocols. We illustrate the use of our framework on two well-known MANET benchmarks: the AODV routing protocol and the leader election protocol of Vasudevan, Kurose, and Towsley. Our formal analysis has uncovered a spurious behavior in the latter protocol that is due to the subtle interplay between communication delays, node movement, and neighbor discovery. This behavior therefore cannot be found by analyses that abstract from node movement and communication delays.

A QoS-Assured and Mobility-Aware Routing Protocol for MANETs

“…Protocols including ZRP, ARPAM, OORP, HSR and CGSR make up this group, which use a combination of the techniques used by reactive and proactive routing protocols, i.e., update packets are sent similar to proactive protocols but with longer intervals and on-demand routing occurs only when there is no viable path from source to destination. These routing protocols are utilized both in wired networks with fixed infrastructure and wireless networks such as MANETs according to the networks' efficiency needs [19][20][21][22][23].…”

Formal Analysis of Leader Election in MANETs Using Real-Time Maude

“…We summarize the Real-Time Maude model of mobility and wireless communication first presented in [13] and improved in our current work (see [15] for more details).…”

“…The status attribute is either pausing or moving, and boundaryTimer denotes the time until the node hits the area boundary. The rewrite rules modeling node movement are given in [15].…”

“…When a node is no longer waiting for any ack message (acks is empty), and it has not yet sent an ack to its parent (pflag is false), it sends an ack message to its parent, with its best-valued node M. However, if the node initiated this round of the protocol (and therefore is the root node) it starts propagating the leader M to its immediate neighbors: Timed Behavior. Due to lack of space, we refer to [15,14] for the definition of the timed behaviors. The main point is that time cannot advance when an instantaneous rule is enabled, so that all actions are performed in a timely manner.…”

Authenticating and Securing Ad-Hoc Networks using Gateway Selection Algorithm

“…Gateway selection can be easily integrated with the routing protocol and provide security [9]. This mechanism is called multi-level authentication technology which includes proactive routing protocols and reactive or hybrid routing protocols.…”