Yurii Kulakov1, professor, D.S., & Mikhail Trifonov1,
post graduate student
1 Computer
System Department, Faculty of Information and Computing Technique, National
Technical University of Ukraine “Kyiv Polytechnic Institute”, Address: 37,
Prospect Permohy, 03056, Kyiv-56, Ukraine, e-mail: mikle@idg-soft.com
Recent
years have witnessed a proliferation of mobile devices. Corporations and
government agencies alike are increasingly using embedded and wireless
technologies, and working towards mobilizing their workforce. Mobile devices
typically support several forms of wireless connectivity like 802.11, IrDA,
Bluetooth, GPRS etc. Due to technology limitations, however, wireless access to
the service providing infrastructure (cell towers, WLAN base-stations) is
limited to particular areas.
Ad
hoc networks, as the name suggests, have no supporting infrastructure. Ad hoc
networks are comprised of a dynamic set of cooperating peers, which share their
wireless capabilities with other similar devices to enable communication with
devices not in direct radio-range of each other, effectively relaying messages
on behalf of others. Conventional methods of identification and authentication
are not available, since the availability of a CA or a KDC cannot be assumed.
Consequently, mobile device identities or their intentions cannot be predetermined
or verified.
Conventional
IDSs have relied on monitoring real time traffic at
switches, gateways, and routers. Vulnerabilities in Medium Access Control (MAC)
for wired networks have been protected by physical
partitioning and restricted connectivity amongst networks. The wireless
connectivity of mobile nodes shares a common medium but cannot be partitioned,
nor can the mobility of the nodes be restricted. Mobility introduces additional
difficulty in setting up a system of nodes cooperating in an
IDS. A node’s movements cannot be restricted in order to let the IDS
cooperate or collect data and a node cannot be expected to monitor the same
physical area for an extended period of time. A single node may be unable to
obtain a large enough sample size of data to accurately diagnose other nodes.
We present interesting results that provide insights into practical
considerations in such a deployment that have not been addressed thus far, and
are not apparent from simulations.