Performance evaluation of random and handshake-based channel access in collaborative mobile underwater networks

In this paper, we compare communication schemes in scenarios of interest for underwater networks where multiple nodes collaborate towards the accomplishment of a target detection and tracking task. We consider two specific cases: in Scenario 1, a mobile Autonomous Underwater Vehicle (AUV) i) collects data from a set of fixed sensors deployed on the seafloor or ii) transmits data to the same nodes. In Scenario 2, two AUVs are placed at opposite sides of an area to be patrolled, and move in the same direction. Some low-complexity, relay-only AUVs keep moving in between edge AUVs in order to support networked data exchange between them. We compare the performance of random as well as handshake-based communications by means of the ALOHA and the DACAP protocols, respectively, in both scenarios. Simulations are performed in a realistic environment, where environmental parameters are extracted from ocean databases and fed to the Bellhop channel simulator through the WOSS framework which seamlessly interfaces the nsMIRACLE network simulator with the more accurate channel representation provided by Bellhop. Results show that while base protocol configurations offer reasonable performance only in scenarios with limited traffic, simple improvements such as back-to-back packet transmission and power control provide significant performance improvements.