Graduate Exam Abstract

Dulanjalie Dhanapala

Ph.D. Preliminary
March 30, 2011, 10.30 am
Rockwell 118
Network-awareness in Wireless Senor Networks - For intelligent organization and efficient data dissemination

Abstract: Wireless sensor network is a densely deployed smart sensor nodes which can self-organize, adapt and collaborate to facilitate distributed monitoring and actuation over large areas. However, due to the resources such as, memory, and energy within the limited capabilities like, communication range and computational power algorithm designing for example, for the information exchange among nodes in these WSNs is a challenge. In general the size of a WSN may vary from hundreds of nodes to thousands in number. Thus, when sensors are deployed, for instance dropped by an air craft, they are clueless about the environment. The only information available to nodes is the number of neighbors and their IDs. But for 2D and 3D sensor network examples, such as Target tracking, habitat monitoring, underground plume tracking, underwater sensor networks, military applications, surveillance applications and health care applications benefit from self- awareness and the network-awareness of the network. GPS, and RSSI based localization schemes are available in literature but facilitating nodes by a GPS is costly and infeasibility for some applications. WSN limitations such as memory, energy and cost constraints are some restrictions against using GPS in WSNs. If a high cost is spent in the organization phase, network will die-out before it performs its actual signed service. Thus investigating energy efficient algorithms and techniques to structure WSN is essential. Other than localization based structuring schemes, clustering algorithms are used for structuring WSNs. Again high initial cost as well as exhausting subset of nodes which called cluster heads shortens the life span of the network. Another available algorithm is Virtual Coordinate Systems (VCSs). VCS is a WSN friendly approach of each node characterizing by a coordinate vector of size M, consisting of distances to each of a set of M anchors. This coordinate system is a higher dimension representation of the actual network. Hop distance based coordinate generation capture the connectivity information of the network while losing the sense of directionality (cardinal or egocentric directions) of the neighborhood. VCS based network structuring schemes will be explored. Network nodes are oblivious. Our aim is to make nodes aware of its neighborhood and its position is the network. Thus exact positioning it not required as far as neighborhood is preserved. Developing techniques and algorithms to achieve reliable and efficient information exchange via self organization of large WSNs in order to achieve node awareness ultimately is the main goal of this proposal. Node awareness is that node knows its neighborhood, position in the network, its own capabilities, limitations, application it is serving and the network wide status of the sensed phenomena. When node awareness is combined with the node mobility, one can envision the future WSNs as, catering more than single application, with dynamically adaptive network structuring based on node’s local decisions to provide better service optimizing resource usage. Advantages of nodes becoming network aware are enormous. Without having any additional hardware just learning from the information disseminated in the network achieving network awareness is not straight forward. Initially properties of VCS, novelty analysis of each anchor, dimension reduction and a transformation to regain the lost directionality will be investigated. Then topology preserving map generation without using any localization scheme will be proposed. Based on the topology preserving maps a novel method of static or dynamic boundary detection scheme will be proposed. Starting from densely deployed oblivious sensor nodes, achieving network awareness from the packets that traverse in the network is also among the contribution of the proposal. Completed work can be summarized as below. A method of evaluating the amount of novel information contained in an ordinate, i.e., in an anchor, on the coordinate space created by the rest of the anchors will be proposed. This method can be used to identify unnecessary or inefficient anchors as well as good anchor locations, and thus help lower overhead and power consumption in routing. Second, a method for reducing the VCS dimensionality is presented. This Singular Value Decomposition (SVD) based method preserves the routability achieved in original coordinate space with lower dimensions. Centralized and online realizations of the proposed algorithm are explained. Then for the first time a novel method of obtaining a topology preserving map from virtual coordinates of a sensor network is presented. Physical layout information such as physical voids and even relative physical positions of sensor nodes with respect to X-Y directions are absent in a VCS description, and obtaining the physical topology has not been possible up to now. A novel technique, based on Singular Value Decomposition, is presented to extract a topology preserving map from VCS. Then extending the topology preserving map generation in 2D to 3D, A method of generating topology preserving maps of networks deployed on 3-D surfaces, based on the VCS, is presented. Proposed method is applicable to wireless sensor networks as well as emerging nano-network applications where nodes are deployed on 3-D surfaces. A topology preserving map of a network of nodes in effect is an identification of the correct embedding out of all the possible embeddings in the higher dimensional virtual domain, which facilitates the identification of network internal and external boundaries without the need for physical location information. A novel boundary identification technique based on the topology preserving maps is presented and evaluated for deployment of network nodes over 2-D and 3D surfaces. Boundary detection scheme presented is simple, not computationally intensive, energy efficient and can be used with physical coordinates as well. After that, a novel technique that combines the advantages of geographic and logical routing called Geo-Logical Routing (GLR), to achieve higher routability at a lower cost is designed. It uses topology domain coordinates, derived solely from virtual coordinates (VCs), as a better alternative for physical location information. By switching between geographic routing and logical routing, GLR overcomes local minima in the respective domains. Next, a novel concept of achieving network awareness at individual nodes in large scale Wireless Sensor Networks (WSNs) is described. With traditional methods, achieving such network awareness requires costly localization integrated with global distribution of location information. In contrast, the proposed scheme achieves network awareness by resorting to passive listening at individual nodes of normal messages associated with applications. Finally, a Directional Virtual Coordinate System (DVCS) is proposed based on a novel transformation that restores the lost directionality information in a Virtual Coordinate System (VCS). The virtual directionality introduced, alleviates the local minima issue present in original VCS. Properties of this virtual directional domain are discussed. With these directional properties, it is possible, for the first time, to consider deterministic algorithms in the virtual domain, as illustrated with a constrained tree network example. A novel routing scheme called Directional Virtual Coordinate Routing (DVCR), which illustrates the effectiveness of the Directional Virtual Coordinate domain, is proposed. VCS based topology map generation in different parts of the network then patching each map together will provide better maps is under investigation. Directional Virtual Coordinate System based topology preserving map generation by selecting nearly orthogonal virtual directions and anchor placement strategy will be addressed. Fault tolerant analysis of the proposed network- awareness for oblivious nodes will be analyzed.

Adviser: Prof. Anura P. Jayasumana
Co-Adviser: N/A
Non-ECE Member: Prof. Indrakshi Ray,Department of Computer Science
Member 3: Prof. Ali Pezeshki,Department of Electrical & Computer Engineering
Addional Members: Prof. Michael Kirby,Department of Mathematics & Department of Computer Science.

Journal Papers 1. D.C. Dhanapala and A.P. Jayasumana, “Convex Subspace Routing (CSR): Routing via Anchor-Based Convex Virtual Subspaces in Sensor Networks,” Computer Communications. Jan. 2011. 2. D.C. Dhanapala, A.P. Jayasumana and Q. Han, “On Random Routing in Wireless Sensor Grids: A Mathematical Model for Rendezvous Probability and Performance Optimization,” Journal of Parallel and Distributed Computing. Nov. 2010 Conference Proceedings: 1. D.C. Dhanapala and A.P. Jayasumana,” Clueless Sensors to Network-Cognizant Smart Sensors: Topology Awareness Via Self-learning in Wireless Sensor Networks” ACM SIGCOMM 2011, Toronto, Ontario, Canada, Aug. 15-19, 2011 (in review) 2. D.C. Dhanapala and A.P. Jayasumana, “Geo-Logical Routing in Wireless Sensor Networks,” Proc. 8th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON), June 2011.(in review) 3. D.C. Dhanapala, S. Mehta and A.P. Jayasumana, “On Topology Mapping and Boundary Detection of Sensor and Nanonetworks Deployed on 2-D and 3-D Surfaces,” Proc. 8th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON), June 2011. (in review) 4. D.C. Dhanapala and A.P. Jayasumana, “Directional Virtual Coordinate System for Wireless Sensor Networks,” Proc. IEEE International Conference on Communications (ICC 2011), June 2011. (accepted) 5. D.C. Dhanapala and A.P. Jayasumana, “Dimension Reduction of Virtual Coordinate Systems in Wireless Sensor Networks,” Proc. IEEE global communications conference (GLOBECOM 2010), Dec. 2010. 6. D.C. Dhanapala and A.P. Jayasumana, “Topology Preserving Maps From Virtual Coordinates for Wireless Sensor Networks,” Proc. 35th Annual IEEE Conference on Local Computer Networks (LCN 2010), Oct. 2010. (Best paper award) 7. D.C. Dhanapala and A.P. Jayasumana, “CSR: Convex Subspace Routing Protocol for WSNs,” Proc. 34th IEEE Conference on Local Computer Networks (LCN 2009), Oct. 2009. 8. D.C. Dhanapala, A.P. Jayasumana and Q. Han, “Performance of Random Routing on Grid-Based Sensor Networks,” Proc. IEEE Consumer Communications and Networking Conference (CCNC) - Personal Ad Hoc and Sensor Networks, Jan. 10 -13, 2009. 9. E.M.N. Ekanayake, M.B. Dissanayake and D.A.D.C. Dhanapala, “MRC Diversity Reception of M-ary DPSK signals in slow Nakagami fading channel,” IEE Electronic Letters, Vol. 42 No. 4, Feb. 16, 2006. 10. E.M.N. Ekanayake, D.A.D.C. Dhanapala and M.B Dissanayake, “EGC Diversity Reception of CPSK Signals in Nakagami Fading,” Proc. IEEE Int. Conference on Industrial and Information Systems (ICIIS 2006), Aug. 08 -12, 2006. 11. D.A.D.C. Dhanapala, M.P.B. Ekanayake, S.S. Gunawardane, A.M. Jayathileke and D.M.D. Jayawickrama; “PC based General Purpose Controller Application Board,” Annual Technical Conference organized by IEE young members section Sri Lanka, Aug. 20, 2005. 12. D. Dhanapala and M. Dissanayake, “Symbol Error Probability Analysis of Nakagami-m Distributed Channel Using Rayleigh Distributed Channel,” Annual Technical Conference organized by IEE young members section Sri Lanka, Aug. 20, 2005.

Program of Study: