The Journal of Systems and Software 84 (2011) 1577–1590 Contents lists available at ScienceDirect The Journal of Systems and Software jo u rn al hom epage: www.elsevier.com/locate/jss A geographic routing hybrid approach for void resolution in wireless sensor networks Gustavo Weber Denardin ∗ , Carlos Henrique Barriquello, Alexandre Campos, Ricardo Nederson do Prado Electronic Ballast Researching Group (GEDRE), Federal University of Santa Maria (UFSM), Santa Maria, RS 97105-900, Brazil a r t i c l e i n f o Article history: Received 31 August 2010 Received in revised form 10 March 2011 Accepted 10 March 2011 Available online 22 March 2011 Keywords: Wireless sensor networks Geographic routing Void resolution a b s t r a c t Geographic routing is one of the most suitable routing strategies for large scale wireless sensor networks due to its low overhead and high scalability features. A geographic routing scheme usually combines a geographic greedy forwarding with a recovery mechanism to solve the local minima problem. Solutions proposed in the literature commonly combine greedy forwarding with the well known face routing for achieving this goal. However, the average path length in number of hops produced by face routing could be much worse than the optimal topological path in most realistic scenarios. In this paper, we propose a new intermediate procedure between the geographic greedy mode and the recovery mode in order to improve routing efficiency in number of hops, without network overhead. It exploits the optimal topological route to base stations, obtained by beacon messages, as a resource to find better routes than the ones created by face routing. We show by simulations that the proposed hybrid approach leads to a significant improvement of routing performance when applied to combined greedy-face routing algorithms. © 2011 Elsevier Inc. All rights reserved. 1. Introduction Wireless sensor networks (WSNs) consist of a large number of densely deployed sensors that have communication, computing, and sensing capacities (Akyildiz et al., 2002). These sensors have the ability to collect, process, and store environmental informa- tion as well as to communicate with others through a wireless channel (Tsai et al., 2009). Such features, combined with its low cost and ease of deployment, make the WSN an attractive technol- ogy for a wide range of applications. These networks have power and memory constraints, low duty cycle, unidirectional data flow (many-to-one flows) and distributed processing. Currently, WSN physical and data link layers are based on well-established standards. Nevertheless, network formation and multi-hop routing is still a challenging problem. A variety of protocols were developed in order to properly perform these rout- ing tasks. However, those existing protocols may not satisfy the requirements for large-scale applications (Sohraby et al., 2007). Geographic routing approach is one of the most suitable rout- ing strategies for such scenario due to its low overhead and high scalability features (Karp and Kung, 2000; Giordano et al., 2001; Subramanian et al., 2007). ∗ Corresponding author. Tel.: +55 55 32209492; fax: +55 55 32208030. E-mail address: gustavo.denardin@gmail.com (G.W. Denardin). Geographic routing mainly relies on a really simple geographic greedy-forwarding strategy, where each router node must select a locally optimal neighbor with a positive progress towards the data packet destination (Chen and Varshney, 2007a). However, when considering realistic outdoor scenarios, such as a city, any block could turn into a void. A void occurs when all neighbors of a router node are farther away from the destination node than the router itself. In this case, the router fails to locate a next hop node through greedy strategy. If only the greedy-forwarding algorithm is used, packets have to be discarded when a void is reached. However, a void does not necessarily mean that there is no route to the des- tination, because if the network is connected, a valid topological path exists by circumventing the void. Many researchers argue that a dense deployment of wireless nodes can reduce the likelihood of the void occurrence in the net- work (Giordano et al., 2001; Al-Karaki and Kamal, 2004; Chen and Varshney, 2007a). Indeed, in scenarios without obstacles that can be true. However, in outdoor large scale wireless networks it is unlikely to have such scenarios. Void handling techniques is an important issue to be addressed in such networks. Although there are several solutions, most of them present some limitations, such as: low effectiveness of han- dling voids (no guarantee of delivery), high resource occupancy, long detours of stuck packets, not optimal path discovery and high complexity. Moreover, most of void-handling techniques are designed with some unrealistic assumptions (Chen and Varshney, 2007a). Among these we can mention equal node transmission 0164-1212/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jss.2011.03.030