200 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 17, NO. 1, FEBRUARY 2009 Virtual Surrounding Face Geocasting in Wireless Ad Hoc and Sensor Networks Jie Lian, Member, IEEE, Yunhao Liu, Senior Member, IEEE, Kshirasagar Naik, Member, IEEE, and Lei Chen, Member, IEEE Abstract—Geocasting in wireless sensor and ad hoc networks means delivering a message from a source node to all the nodes in a given geographical region. The objectives of a geocasting protocol are two-fold: guaranteed message delivery and low transmission cost. Most of the existing protocols do not guarantee message de- livery, and those that do, incur high transmission costs. In this study, we propose the concept of Virtual Surrounding Face (VSF), and design a VSF-based geocasting protocol (VSFG). We also design a SKIP method and a local dominating set (DS) based restricted flooding technique to further reduce the cost of VSFG. Through mathematical analysis and comprehensive sim- ulations, we show that VSFG, together with SKIP and local DS based restricted flooding, guarantees message delivery and has a much lower transmission cost than the previous approaches. The reduction of cost can be up to 65% compared with the most effi- cient existing approach. Index Terms—Ad hoc networks, geocasting, virtual surrounding face, wireless sensor networks. I. INTRODUCTION G EOCASTING in wireless sensor network is a task to de- liver a message from a source node to all nodes located within a given geographic region. An important objective of geocasting is to ensure message delivery while maintaining a low transmission cost (lower number of transmissions). Guar- anteed delivery ensures that every sensor in a region receives a copy of the geocasting message. Since sensors are generally powered by batteries, the limited energy of sensors requires geocasting to consume as little energy as possible. Many algo- rithms have been proposed in the literature [9]–[18] to achieve geocasting. The approaches presented in [9]–[16] do not guar- antee message delivery and incur high transmission costs. Of the existing approaches, four algorithms—one in [17] and three in [18]—guarantee message delivery in continuous geocasting Manuscript received December 30, 2006; revised August 27, 2007 and November 30, 2007; approved by IEEE/ACM TRANSACTIONS ON NETWORKING Editor S. Das. First published July 25, 2008; current version published Feb- ruary 19, 2009. This work was supported in part by the NSERC (Canada), the National Basic Research Program of China (973 Program) under Grant 2006CB303000, the National High Technology Research and Development Program of China (863 Program) under Grant 2007AA01Z180, NSFC Key Project Grant 60533110 and Grant 60736016, the Hong Kong RGC Grant HKUST6169/07E and Grant N_HKUST614/07, and the HKUST Nansha Research Fund NRC06/07.EG01. A preliminary, shorter version of this paper was presented at the IEEE ICNP, November 2006. J. Lian and K. Naik are with the Electrical and Computer Engineering De- partment, University of Waterloo, Waterloo, ON, N2L 3G1 Canada (e-mail: jlian@swen.uwaterloo.ca; knaik@swen.uwaterloo.ca). Y. Liu and L. Chen are with the Computer Science Department, Hong Kong University of Science and Technology, Hong Kong, China (e-mail: liu@cse. ust.hk; leichen@cs.ust.hk). Digital Object Identifier 10.1109/TNET.2008.927251 Fig. 1. Inefficiency of face traversal in the existing approaches. In the figure, the dotted circle denotes the transmission radius of . regions. Those algorithms, however, incur high transmission costs. In this paper, we propose a geocasting algorithm based on the idea of Virtual Surrounding Face (VSF), and we refer to this algorithm as VSF Geocasting (VSFG). We prove that VSFG guarantees message delivery to the nodes within a geocasting region. In addition, the transmission cost of VSFG is signifi- cantly reduced compared with the existing approaches. Guaran- teed message delivery in a connected network means that the message can be delivered from any source to any destination with an assumption: for any two neighbor nodes, a MAC layer protocol, such as 802.11a/b and [25], [27], exists to guarantee correct message exchange between them. In VSFG, a network topology is converted into a planar graph where no two edges cross one another. The network area is partitioned into a set of faces, where a face is a continuous area enclosed by a sequence of edges. In VSFG, all the faces intersecting with a geocasting region are merged into a unique virtual surrounding face containing . VSFG includes the following three steps: VSF forwarding, VSF traversal, and restricted flooding. In VSF forwarding, a source delivers a geocasting message to a node on the boundary of VSF, called a VSF node, by using location-based routing [3], [4]. In VSF traversal, the VSF node initiates a face traversal in which all the nodes on the VSF receive a copy of the message. Finally, in restricted flooding, nodes in that overhear the face traversal message perform restricted flooding within . Many approaches [3], [4], [18] can be used for face traversal. Those approaches, however, are not efficient in terms of mes- sage complexity. As illustrated in Fig. 1, node v starts a face traversal along the paths v w x y . In the ex- isting approaches, even though x is a direct neighbor of v, the message is sent from v to w, and then from w to x, introducing an extra transmission. One intuition is that in dense networks, these additional transmissions may be significant compared with the total number of transmissions for face traversal. To reduce the 1063-6692/$25.00 © 2008 IEEE Authorized licensed use limited to: Hong Kong University of Science and Technology. Downloaded on September 16, 2009 at 02:28 from IEEE Xplore. Restrictions apply. This is the Pre-Published Version