Topology and Routing Design for Three-Dimensional Sensor Networks: Geometric Approaches

Sept. 2007 ~ August 2010, National Science Foundation (NSF)

Investigator: Yu Wang;

Student:  Fan Li, Siyuan Chen

Most existing wireless sensor systems and protocols are based on two-dimensional design, where all sensors are distributed in a two dimensional plane. This assumption is somewhat justified for applications where sensors are deployed on earth surface and where the height of the network is smaller than transmission radius of a sensor. However, 2D assumption may no longer be valid if a sensor network is deployed in space, atmosphere, or ocean, where nodes of a network are distributed over a 3D space and the differences in the third dimension is too large to ignore. This project focuses on designing novel geometric approaches to solve various topology control and position-based routing problems in 3D sensor networks. Although many geometric topology control protocols and position-based protocols have been studied in 2D sensor networks, the design of 3D networks is surprisingly more difficult than the design in 2D. Many properties of the network require additional computational complexity, and a number of problems cannot be solved by extensions or generalizations of 2D methods. Facing up with these challenges, this project seeks to study new geometric approaches for 3D sensor networks. The expected results include: (1) various localized algorithms to efficiently construct 3D geometric topologies in order to maintain network connectivity, conserve energy and enable energy efficient routing; (2) new 3D position-based routing methods which can guarantee the delivery of packets or the power efficiency of their routes; (3) integrated 3D geometric approaches to address the joint design of topology and routing where these two issues are strongly coupled and fundamentally influenced by geometry.

A Microeconomic Approach: Protocol Design for Ubiquitous and Integrated Networks

Sept. 2006 ~ August 2007, Oak Ridge Associated Universities

Investigator: Yu Wang;

Student:  Fan Li

A ubiquitous and integrated architecture has been envisioned for future networking. Most of the algorithms and protocols designed in computer network implicitly assume that the participating computers/users will act as instructed. However, end users in a wireless network are interested in the share of the radio spectrum they enjoy, not in the global optimum of the system. If these nodes act selfishly (e.g., refuse to relay the data while the routing protocol assumed that they will), it may hinder the functioning of the network. Therefore, protocols intended for selfish computers/users must be designed in advance to cope with selfishness for future ubiquitous and integrated networks. In this project, we investigate the impact of users' selfish behavior on the performance of several particular network protocols in integrated networks, and aims to design truthful network protocols based on game theory approaches for ubiquitous and integrated networks with various selfish participants.

Dynamic Topology Control for Wireless Sensor Networks

July 2006 ~ December 2007, University of North Carolina at Charlotte, Faculty Research Grant (FRG)

Investigator: Yu Wang;

Student:  Fan Li

Unlike wired networks, in wireless sensor networks, each wireless node can move and thus change the topology of the network. In this case, we need to adjust the transmission power or selected neighbors to keep some properties of the network topology such as connectivity or power efficiency. On the other hand, flexible topologies are enabled by multiple simultaneous frequencies to multiple adjacent nodes in wireless sensor networks. Therefore, we need to efficiently determine how to choose among all possible topologies and how to evaluate network architectures. To address these problems, dynamic topology optimization (also called topology control) is emerging as one of the critical issues in the implementation of wireless sensor networks. The main goal of this projectl is to develop, design, implement and test distributed topology control protocols for wireless sensor networks.

Experimental Testbed for Mobile Network Protocols

Sept. 2001 – Jan. 2004, NSF/CISE-RR 013799

Investigators: Teresa Dahlberg, Essam Elkwae, Gail Ahn, Asis Nasipuri

Students:  Vinod Namboodiri, Shirisha Thummala

The overall objective of this project is to experimentally analyze mobile network protocols that support multimedia services.  A wireless, mobile multimedia network is being built to add an experimental component to four ongoing research projects at UNC Charlotte.  The experimental work will focus on the component of each project that involves development and analysis of mobile network protocols.  Experimentation will enable critical analysis of protocol behavior in dynamic environments where real-world entities replace simulation models, especially, network traffic models, wireless channel models, fault and vulnerability models, and power usage models.  The testbed will encompass both cellular and ad hoc network architectures with components that include PCs and laptops with IEEE 802.11 radios and FreeBSD operating system.  Network nodes to be configured include multimedia nodes that generate variable bit rate streaming audio and video and a security authentication node.