University of Houston

Department of Computer Science

In Partial Fulfillment of the Requirements for the Degree of
Master of Science

Sumit Gupta

Will defend his thesis

Automatic Detection of DOS Routing Attacks in

Wireless Sensor Networks

Abstract

The development of wireless sensors which are able to communicate their data to base stations without a wire link have accelerated the use of sensors in real world at lower cost. Despite their many advantages, the design of wireless sensor networks (WSNs) is subject to a few practical challenges such as limited energy budget, constrained processing and storage capacity and security vulnerabilities.

WSNs are highly prone to security threats, and hence there is a need of some intrusion detection system which can assist the network administrator in detecting and root-causing the launched attacks. This thesis proposes the concept of Sensor Network Automated Intrusion Detection System (SNAIDS), a detection system that has been designed to automate the attack detection process in WSNs, and can be managed and monitored from centralized location(s). Unlike many other existing IDSs, which use decentralized detection algorithms, SNAIDS employs a centralized detection solution which imposes less overhead on sensor nodes with better accuracy. Also, in contrast to other existing centralized approaches, SNAIDS uses a dual routing layer, which makes it even harder for intruders to compromise it completely.

We have implemented a prototype of SNAIDS using nesC and Java on platform, and evaluated its performance on a small 22-node sensor network test-bed against black hole, sink hole, selective forwarding, flooding and misdirection attacks. Experiment studies show that the proposed IDS is capable of detecting all the attacks in most testing scenarios.

Date: Wednesday, November 29, 2006
Time: 1:00 PM
Place: 362-PGH

Faculty, students, and the general public are invited.
Advisors: Dr. Albert M. K. Cheng, Dr. Rong Zheng