3.4 Performance Evaluation
We evaluate the performance of EGOR through extensive Monte-Carlo simulations. We compare EGOR with geographic routing, which only has one forwarding candidate that achieves the maximum EPA, and opportunistic routing, which involves all the available next-hop nodes as forwarding candidates. When the packet becomes stuck, all the three protocols just drop the packet. Various situations are simulated by varying node densities, transmission-to-reception energy ratios and retransmission limits.
3.4.1 Simulation Setup
Channel Model: To simulate a realistic channel model for lossy WSNs, we use the log-normal shadowing path loss model derived in (Zuniga and Krishnamachari 2004):
where d is the transmitter–receiver distance, γ(d) is the signal-to-noise ratio (SNR), ρ is the encoding ratio and Lf is the frame length in bytes. This model considers several environmental and radio parameters,3 such as the path-loss exponent (α) and log-normal shadowing variance (σ) of the environment, and the modulation and encoding schemes of the radio. This particular equation resembles a MICA2 mote see the MICA2 datasheet at www.eol.ucar.edu/rtf/facilities/isa/internal/CrossBow/DataSheets/mica2.pdf, which has data rate of 19.2 kbps, and the noise bandwidth 30 kHz. Non-coherent FSK and Manchester are used as the modulation and encoding schemes (ρ = 2), respectively. The environmental ...