Chapter 3
Energy Efficiency of Geographic Opportunistic Routing
Wireless sensor networks (WSNs) are characterized by multihop lossy wireless links and severely resource-constrained nodes. Among the resource constraints, energy is probably the most crucial one since sensor nodes are typically battery powered and the lifetime of the battery imposes a limitation on the operation hours of the sensor network. Unlike the microprocessor industry or the communication hardware industry, where computation capability or the line rate has been continuously improved (regularly doubled every 18 months), battery technology has been relatively unchanged for many years. Energy efficiency has been a critical concern in wireless sensor network protocol design. Researchers are investigating energy conservation at every layer in the traditional protocol stack, from the physical layer up to the network layer and application layer.
Among the energy consumption factors, communication has been identified as the major source of energy consumption and costs significantly more than computation in WSNs (Pottie and Kaiser 2000). Opportunistic routing has shown its advantage on energy efficiency (Zeng et al. 2007; Zorzi and Rao 2003) comparing to traditional routing. However, the existing opportunistic routing schemes like GeRaF (Zorzi and Rao 2003) typically include all the available next-hop neighbors as forwarding candidates, which does not lead to optimal energy efficiency.
In this chapter, we propose an energy-efficient geographic opportunistic routing (EGOR) framework, which is based on opportunistic routing but more judiciously selects a subset of the available next-hop neighbors as the forwarding candidates to strike a good balance between the packet advancement and energy cost. The analysis of how to achieve maximum EPA in Chapter 2 provides useful insights on the selection of the forwarding candidate set. Based on which, we propose two localized candidate selection algorithms with O(M3) and O(M2) running time in the worst case respectively and Ω(M) in the best case, where M is the number of available next-hop neighbors of the transmitter. The algorithms efficiently determine the optimal forwarding candidate set with respect to the EPA per unit of energy consumption. The performance of EGOR is justified through extensive simulations and comparisons with those of the existing geographic routing and opportunistic routing schemes. The simulation results show that EGOR strikes a good balance between energy consumption and routing efficiency in terms of EPA, and achieves the best energy efficiency among the three schemes in all the cases. Our simulation results also show that, under a realistic lossy channel model, the best energy efficiency can be achieved with only a very small number of forwarding candidates (around two), even when the energy consumption of reception is negligible to that of transmission.
The rest of the chapter is organized as follows. We formulate the EGOR problem in Section 3.1. Two efficient localized candidate selection algorithms are proposed in Section 3.2. In Section 3.3, we propose and analyze our EGOR scheme. Simulation results are presented in Section 3.4. Conclusions are drawn in Section 3.5.