Channel replay-based performance evaluation of protocols for underwater routing

Abstract

The paper presents a comparative performance evaluation of two routing protocols for underwater wireless sensor networks (UWSNs). The two protocols are the ChannelAware Routing Protocol (CARP), exemplary of a cross layer approach to underwater routing, and a simple variation of common flooding, called EFlood, where performance is enhanced by introducing random re-transmission times. The scenarios we consider are obtained via simulations and from trials at sea performed under a collaboration agreement between the University of Roma "La Sapienza" and the NATO Science and Technology Organization Centre for Maritime Research and Experimentation (STO CMRE). Two sets of simulations results are shown where the physical layer is modeled by a ray-traced channel as well as by replaying real channel traces, under the same network configuration. Results are also reported from campaigns of experiments at sea. Comparing all results shows how channel replay mimics faithfully actual channel dynamics with respect to what is achievable through a simulated channel model, thus demonstrating the effectiveness of this technique for a fair and repeatable performance comparison of solutions for UWSNs.

The SUNSET framework for simulation, emulation and at-sea testing of underwater wireless sensor networks

“…Simulations using different underwater acoustic channel models, such as empirical formulas (e.g., Urick's models [31]), the Bellhop [32] propagation simulator via the WOSS [33] interface (which includes real environmental parameters and provide more accurate results), and traces of the channel used for channel replay purposes [34]. Reuse of the same code for simulation, emulation and at-sea test mode, introducing little additional delays (in the order of milliseconds on very low-power embedded devices) with limited additional overhead per packet (few bits).…”

“…Two SUNSET modules have been designed to allow the use of channel replay techniques [34,39]. The first module provides functionalities for collecting channel quality information by triggering periodic packet transmissions by each node.…”

“…Additionally, SUNSET was also used to collect channel measurements about the quality of the links between pairs of nodes, which were then used to evaluate the performance of the different routing protocols by means of channel replay-based simulations [34].…”

MPR

“…A detailed description on the processing of the collected information and on the use of channel replaying with SUNSET can be found in [10].…”

“…The common flooding protocol, where each node immediately retransmits a received packet (unless it is a duplicate), does not incur in these kind of issues but it results in a large number of collisions in UASNs. EFlood [10] enhances the common flooding by letting a node wait for a random time before forwarding the packet. This allows EFlood to reduce the probability of collision when multiple nodes forward the same packet.…”