DOCOMO Labs United States
DOCOMO Labs United States
Saipulla A.,University of Massachusetts Lowell |
Westphal C.,DOCOMO Labs United States |
Liu B.,University of Massachusetts Lowell |
Wang J.,University of Massachusetts Lowell
Ad Hoc Networks | Year: 2013
Barrier coverage of a wireless sensor network is a critical issue in military and homeland security applications, aiming to detect intruders that attempt to cross the deployed region. While a range of problems related to barrier coverage have been investigated, little effort has been made to explore the effects of different sensor deployment strategies and mechanisms to improve barrier coverage of a wireless sensor network after it is deployed. In this paper we study the barrier coverage of a line-based sensor deployment strategy and explore how to exploit sensor mobility to improve barrier coverage. We first establish a tight lower bound for the existence of barrier coverage under the line-based deployment. Our results show that the barrier coverage of the line-based deployment significantly outperforms that of the Poisson model when the random offsets are relatively small compared to the sensor's sensing range. To take advantage of the performance of line-based deployment, we further devise an efficient algorithm to relocate mobile sensors based on the deployed line so as to improve barrier coverage. The algorithm finds barrier gaps and then relocates mobile sensors to fill the gaps while at the same time balancing the energy consumption among mobile sensors. Simulation results show that the algorithms can effectively improve the barrier coverage of a wireless sensor network for a wide range of deployment parameters. Therefore, in wireless sensor network applications, the coverage goal, possible sensor deployment strategies, and sensor mobility must be carefully and jointly considered. The results obtained in this paper will provide important guidelines and insights into the deployment and performance of wireless sensor networks for barrier coverage. © 2011 Elsevier B.V. All rights reserved.
Westphal C.,Docomo Labs United States |
Perkins C.E.,WiChorus Inc.
IEEE International Conference on Communications | Year: 2010
The number of devices connected to the Internet has outstripped the number of effectively assignable IPv4 addresses. In order to be globally reachable, many devices must share the same IPv4 address; current mechanisms only provide reachability when the device sharing the IPv4 address itself initiates communication. We describe a mechanism to make nodes behind a NAT globally reachable, even when communications are initiated from the global Internet. The intended application of the mechanism, denoted SIPNAT , is to allow for the first time bidirectional global reachability of IPv6 addresses by nodes in the global IPv4 Internet, in a scalable manner, thus resolving the major issue associated with IPv4-IPv6 translation. SIPNAT involves filtering flows at the gateway between the IPv4 and IPv6 domain through a combination of DNS request and timing information for the IPv4 initiated connection (the IPv6 to IPv4 connections are performed using typical NAT mechanisms, where the IPv6 domain takes the role of the private address space). We study the performance of the SIPNAT mechanism using queuing theoretic analysis, and show that our SIPNAT model is accurate on actual data traces. ©2010 IEEE.