Performance Comparison of Opportunistic Routing Schemes in Wireless Sensor Networks Petros Spachos, Liang Song, and Dimitrios Hatzinakos Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON M5S 3G4, Canada E-mail: {petros,songl,dimitris}@comm.utoronto.ca Abstract—Opportunistic routing is considered a promising direction for improving the performance of wireless sensor networks (WSN). In opportunistic routing, the intermediate nodes collaborate on packet forwarding in order to achieve high throughput in the face of lossy links. That makes the next node selection process crucial. In this paper, we are presenting an opportunistic routing protocol for wireless sensor networks. We also examine three extensions of that protocol, based on different next node selection criterion. We illustrate how each extension works and we evaluate and compare their performance in terms of energy consumption, delivery ratio and packet latency. I. INTRODUCTION Wireless sensor networks have quickly gained popularity. Recent advancement in wireless communications and elec- tronics has enabled the development of low-cost, low-power, multifunctional sensor nodes that are small in size. However, the unique characteristics of sensor nodes and their wireless communication can pose significant challenges. Energy consumption is the major challenge in every sensor network. Usually, sensor networks are designed to operate unattended for long periods of time because battery replace- ment or rechargeability is sometimes infeasible or impossible. Therefore, the battery charge must be conserved to extend the life of each sensor and the entire sensor network. When applying a routing scheme, the impact that this scheme has on the lifespan of the network should always be considered. The energy required to transmit related routing data should always be considered when implementing a protocol within a sensor node. The communication range of sensor nodes is also limited in order to conserve energy. A great reduction in the transmission power can save sensor node energy. However, it reduces each sensor node detection probability and communi- cation range. On the other hand, wireless medium may lead to the packets damage or loss due to channel errors. Wireless communi- cations are facing many unpredicted challenges such as air interference, channel fading, environmental changes etc. A reliable routing protocol should handle appropriately any lost or missing packets. Even if the channel is reliable enough, the broadcast nature of wireless communication causes one more problem. If multiple packets meet in the middle of transfer, conflicts will occur and the transfer will fail. In a large scale network with high traffic volume this can be a major problem. Opportunistic routing tries to overcome the drawback of an unreliable wireless link by taking advantage of the broadcast nature of the wireless medium such that one transmission can be overheard by multiple nodes. A cluster of nodes serves as a relay candidate but only one node finally will forward the packet. An opportunistic routing protocol can use different nodes and follow multiple paths toward the destination for each packet transmission. The next node selection process is crucial and based on opportunistic rules. The next node selection criterion of an opportunistic routing protocol, can be defined in order to enhance network security in terms of source-location privacy [1]. In this paper, we are examining the impact of the next node selection criterion on the network performance in terms of energy consumption and network transmission reliability. The rest of this paper is organized as follows. In Section II, the related works are reviewed. The module design and implementation are presented in Section III while opportunis- tic routing extensions are described in Section IV. In Section V, performance analysis and simulation results are presented, followed by conclusions in Section VI. II. RELATED WORK During the last decade, a number of opportunistic protocols have been developed. The first opportunistic routing has been introduced in [2]. Extremely Opportunistic Routing (ExOR) selects the next relay node by a slotted ACK (acknowledge) mechanism. Having successfully received a data packet, the node calculates a priority level, which is inversely proportion- ate to the expected transmission count metric (ETX), [3] , which is based on the distance between the node and the des- tination. The shortest the distance, the highest the priority. The node with the highest priority will then be selected as the next relay node. The main drawback of ExOR is that it prevents spatial reuse because it needs global coordination among the candidate nodes. Candidate nodes transmit in order, only one node is allowed to transmit at any given time while all the other candidate nodes trying to overhear the transmission in order to learn which node will be the next relay node. Moreover, the simple priority criteria that it uses, ETX distance, may lead packets toward the destination through low-quality routes. To overcome this problem, Opportunistic Any-Path Forwarding (OAPF) [4] introduces an expected any-path count (EAX) metric. This can calculate the near-optimal candidate set at each potential relay node to reach the destination. However, it needs more state information about the network and it has high computational complexity. ExOR ties the MAC with routing, imposing a strict schedule on routers access to the medium. The scheduler goes in rounds. 2011 Ninth Annual Communication Networks and Services Research Conference 978-0-7695-4393-2/11 $26.00 © 2011 IEEE DOI 10.1109/CNSR.2011.46 271