English abstract
Underwater Wireless Sensor Networks (UWSNs) are promising for discovering the aqueous environment. They have attracted the interest of many researchers in the last decade. UWSNs are used for military and non-military applications such as undersea exploration of natural resources, tactical supervision, and mines detection. Several research problems related to UWSNs have been tackled. In particular, routing is a challenging problem in UWSNs. Routing protocols for UWSNS should take into consideration several constraints related to the difficult underwater conditions including dynamic topology, limited energy, low bandwidth and high propagation delay of acoustic signals used for communication underwater. In [31] a new Tree-Based Routing protocol called TBR for Underwater Wireless Sensor Networks has been proposed. The protocol constructs shortest-path trees in a virtual 3D grid topology and uses them to obtain routing paths between sensor nodes and sink nodes in the UWSN. It has been claimed that due to the availability of these pre-constructed shortest-path trees, TBR is expected to outperform other protocols which rely on costly reactive mechanisms for path establishment and maintenance. The aim of this project is to confirm this expectation through implementing the TBR protocol over 3D grid for UWSN, evaluating its performance, and comparing it to the performance of other protocols using simulation. The performance measures: delivery ratio, communication delay, and energy consumption have been used in the performance evaluation. The evaluation was conducted by varying the network parameters: density, mobility and load.
The simulation was conducted using Aqua-Sim which is a network simulator for underwater sensor networks. Aqua-Sim is based on NS-2, one of the most widely used network simulators. The obtained results show that TBR outperforms the well-known Vector-Based Forwarding (VBF) routing protocol for UWSNs in terms of packet delivery ratio and average end to end delay. TBR has performed more than two times better than VBF in terms of these two metrics. However, TBR consumed considerably more energy than VBF (up to ten times more in some of the tested cases). It was also observed from the simulations that TBR performs better under lower mobility speeds of the sensor nodes.
