David J. Dorsey

Analysis of Slotted Aloha with multipacket messages in clustered surveillance networks

This work presents an analysis of a cluster of finite population of low cost sensor nodes operating in a p-persistent S-Aloha framework with multipacket messages. Using this analytical framework, we consider the issue of partitioning the nodes and available frequencies into groups so as to maximize the system throughput. Assigning the nodes and frequencies into “groups” is important because the size of the group impacts the tradeoff between the benefits of frequency diversity and the cost of collision on the shared medium imposed by the nodes in a group. We study this tradeoff through analytical and numerical results and show how the correct choice of group sizes can vary depending on various factors like the ratio of nodes to frequencies and the overall system load.

A secure wireless agent-based testbed

Secure mobile wireless ad-hoc networks are frequently described in the technical literature as highly-desired and feasible. However only few stable and scalable physical realizations of such MANETs were actually reported. Here, we describe SWAT (a secure wireless agent-based testbed), a physical network of handheld and stationary computing nodes that provide a practical industrial-strength MANET based on IEEE standard 802.11b and agent technology. The system, using HP iPAQ units and tablet PCs, implements secure group-based applications and allows for real-time user revocation. It integrates algorithms and techniques from several proposed software libraries, including CLIQUES, spread and secure spread, SEM, IPSec, and EMAA.

A framework for energy-efficient adaptive jamming of adversarial communications

This work proposes a framework for jamming wireless networks that incorporates probabilistic models of internal states and observable characteristics of link protocols, where protocols are divided into two general classes: random access (RA) or channelized access (CA). Without exact knowledge of network parameters and internal state, the proposed intelligent jammer optimizes its strategy to be energy efficient while achieving the target throughput. Probabilistic models for jamming FDMA and CSMA-based protocols are described for illustration of the framework: A frequency-hopping voice network is analyzed to determine the optimal jam strategy for proactive frequency jammer; and a CSMA packet protocol is analyzed for varying packet arrival rates at the nodes. Since RA protocols display observable reaction to channel conditions, we propose a feedback-control loop that uses observable feedback to infer network parameters. Both protocols are evaluated through simulation for their energy-throughput tradeoff compared to a naive jammer.

Non-uniform deployment of nodes in clustered wireless sensor networks

In this paper, we consider the problem of placing, with course-grained control, a large number of wireless networked sensor nodes employing a clustering network architecture. The goal of the deployment strategy is to maximize the lifetime while ensuring connectivity between cluster-heads so that samples from the monitored area may be forwarded to a fusion center. A model is derived to approximate the lifetime of a differentially deployed random network using both the density of cluster-heads and non-clusterheads as variables. Through optimization of the lifetime expression over both variables and through simulation results, it is shown that 1) a differential node deployment with a uniform cluster-head density increases the lifetime of the network over a uniform deployment and 2) the addition of a suitable differential cluster-head density further increases lifetime over the differential node deployment with uniform cluster-head density.

Self-Adaptive Dissemination of Data in Dynamic Sensor Networks

The distribution of data in large dynamic wireless sensor networks presents a difficult problem due to node mobility, link failures, and traffic congestion. In this paper, we propose a framework for adaptive flooding protocols suitable for disseminating data in large-scale dynamic networks without a central controlling entity. The framework consists of cooperating mobile agents and a reinforcement learning component with function approximation and state generalization. A component for agent coordination is provided, as well as rules for agent replication, mutation, and annihilation. We examine the adaptability of this framework to a data dissemination problem in a simulation experiment.

Ternary trust metric for mobile ad-hoc networks

In this paper we describe a behavioral state classifier employing a three-factor node reliability measure based upon inferred Ability, Integrity, and Benevolence (AIB) to assess node reliability. In contrast to typical scalar measures, this multi-metric index creates a reliability space that is more descriptive and in which distinctions can be drawn between node behaviors ranging from small deviations in expected performance to signatures of hostile intent. We further demonstrate that a reliability or trust metric should be informative enough to differentiate between malicious behavior and behavior exhibited by nodes that are misconfigured or dropping packets due to insufficient resources to service messages.

Optimal deployment and replenishment of monitoring wireless sensor networks

Wireless Sensor Networks (WSNs) composed of inexpensive devices allow the possibility of large deployments aimed at monitoring remote, or possibly hostile, areas. However, due to the unique traffic patterns exhibited by monitoring sensor networks, the lifetime of a large WSN is constrained by the burden placed on nodes near the sink node to forward additional traffic as more nodes are deployed. We discuss approaches for deploying a WSN that will maximize the lifetime of an initial deployment. We then consider cases where the mission lifetime is of a duration such that overdeploying an initial network to meet this mission criteria would become prohibitively expensive. We then propose a replenishment control framework where additional nodes are added to an initial deployment in consecutive batches in order to meet mission lifetimes while reducing cost. The control framework consists of a failure process model used to forecast sensor failures due to energy depletion, and a two-stage limited lookahead controller used to determine the number of nodes to be added to the network and the approximate locations of their deployment.