Gossiping-based AODV for Wireless Sensor Networks Stefano Galzarano *† , Claudio Savaglio † , Antonio Liotta * and Giancarlo Fortino † * Department of Electrical Engineering Eindhoven University of Technology (TU/e), Eindhoven, The Netherlands Email: s.galzarano@tue.nl, a.liotta@tue.nl † Department of Informatics, Modelling, Electronics and Systems (DIMES) University of Calabria (UNICAL), Rende, Italy Email: g.fortino@unical.it, csavaglio@deis.unical.it Abstract—Wireless sensor networks have been widely used in many different applications and in the future they will play an increasingly important role. Since these networks have no fixed infrastructure and are usually distributed over large areas, the use of routing protocols is indispensable. However, when the number of nodes within an area increases, the communication interferences and collisions increase significantly, thus reducing the network performance. In this paper, we first introduce a new measurable quantity, the ”node concentration”, in contrast to the standard network density. Then, the performance of the AODV (Ad-hoc On-demand Distance Vector) routing protocol is evaluated with respect to the variation in node concentration. Finally, we propose an enhancement of AODV, called CG-AODV, by introducing a ”node concentration-driven gossiping” approach for limiting the flooding of control packets. The simulation results demonstrate that CG-AODV provides significant improvements in terms of packet delivery ratio and path discovery delay. Index Terms—wireless sensor networks, reactive routing, ad- hoc on-demand distance vector, gossiping, traffic control, node concentration, network density. I. I NTRODUCTION A Wireless Sensor Network (WSN) consists of a large num- ber of small, low-cost, and low-power devices providing data acquisition, processing and wireless communication capabili- ties [1]–[5]. WSNs have captured the attention of an increasing number of researchers in diverse disciplines and have proved great potential in many scenarios, such as environment and building monitoring, industrial control, intelligent agriculture, and catastrophic management, among others. Sensor nodes are usually spatially distributed over a certain geographical area and due to the scarce wireless communication range, they need to be organized into a multi-hop network with no fixed infrastructure. Thus, routing represent an indispensable part of such a network and takes on the responsibility to assist sensor nodes in communicating and cooperating with each others for the benefit of the high-level distributed applications. Differently from traditional distributed systems, routing in WSNs is very challenging due to the inherent characteristics and restrictions that distinguish such networks, mainly energy and communication bandwidth constraints, scarce node capa- bilities, absence of IP-based addressing scheme. Moreover, routing is highly influenced by several factors such as node deployment, data delivery models, node/link heterogeneity, connectivity, and coverage. Although available routing protocols may be classified under different points of view [6], they can be generally in- cluded into three main categories: proactive protocols, reactive protocols and hybrid protocols. Research on routing protocol shows that reactive routing is generally preferred to proactive solutions in different aspects, such as network lifetime, self- organizing network model and the load of the network [7], [8]. However, it is well-known that under frequent network topology changes, reactive protocols may suffer from large volume of messaging overhead due to the many necessary route discoveries. But, differently from the MANETs (Mobile Ad-hoc Networks), where routing information may change frequently due to node mobility, in most application scenarios the nodes constituting a WSN are mostly stationary after their deployment. In such cases, a proactive protocol would represent the most adequate solution for packets routing, as it does not need continuous routing information updates; also path discovery requests are not so frequent as long as routes caching is adopted. However, even with stationary nodes, reactive protocols are mostly influenced by other factors, such as network density. In fact, differently from other networks (e.g. MANETs), sensor nodes may need to be densely deployed over a certain area. Network density is usually measured as the number of nodes per unit area, and sometimes it is also referred to as just the number of nodes constituting the network over a certain deployment area. But such a measure in not always suffi- cient to properly characterize a network, because the actual transmission range of the nodes is not explicitly taken into consideration. In fact, collisions and packet overheads in a specific zone of the network are mostly influenced by the concentration of a neighborhood (i.e. the subset of nodes able to directly communicate with each others) rather than the geographical density of the network intended as the number of nodes per square meter. In this paper we provide a two-fold contribution. First, we introduce the concept of ”node concentration”, which considers the average number of neighbors connected to each node, given a specific transmission power (or transmission range). We are interested in investigating how the control 978-1-4799-0652-9/13/$31.00 c 2013 IEEE