No abstract

Simulation forms an important part of the development and empirical evaluation of underwater acoustic network (UAN) protocols. The key feature of a credible network simulation model is a realistic channel model. A common approach to simulating realistic underwater acoustic (UWA) channels is by using specialised beam tracing software such as BELLHOP. However, BELLHOP and similar modeling software typically require knowledge of ocean acoustics and a substantial programming effort from UAN protocol designers to integrate it into their research. This paper is a distilled tutorial on UWA channel modeling with a focus on network simulation, providing a trade-off between the flexibility of low level channel modeling via beam tracing and the convenience of automated channel modeling, e.g. via the World Ocean Simulation System (WOSS). The tutorial is accompanied by our MATLAB simulation code that interfaces with BELLHOP to produce channel data for UAN simulations. As part of the tutorial, we describe two methods of incorporating such channel data into network simulations, including a case study for each of them: 1) directly importing the data as a look-up table, 2) using the data to create a statistical channel model. The primary aim of this paper is to provide a useful learning resource and modeling tool for UAN protocol researchers. Initial insights into the UAN protocol design and performance provided by the statistical channel modeling approach presented in this paper demonstrate its potential as a powerful modeling tool for future UAN research.

Underwater acoustic sensor networks are an enabling technology for many applications. Long propagation delays and limited bandwidth of the acoustic channel place constraints on the trade-off between achievable end-to-end delay, channel utilization, and fairness. This paper provides new insights into the use of the combined free/demand assignment multiple access (CFDAMA) schemes. The CFDAMA can be classified as adaptive TDMA, where capacity is usually assigned on demand. The CFDAMA with round robin requests (CFDAMA-RRs) are shown to minimize end-to-end delay and maximize channel utilization underwater. It sustains fairness between nodes with minimum overhead and adapts to changes in the underwater channel and time-varying traffic requirements. However, its performance is heavily dependent on the network size. The major contribution of this paper is a new scheme employing the round robin request strategy in a systematic manner (CFDAMA-SRR). Comprehensive event-driven Riverbed simulations of a network deployed on the sea bed show that the CFDAMA-SRR outperforms its underlying scheme, CFDAMA-RR, especially when sensor nodes are widely spread. Considering node locations, the novel scheme has a bias against long delay demand assigned slots to enhance the performance of the CFDAMA-RR. The illustrative examples show good agreement between the analytical and simulation results. INDEX TERMS CFDAMA, medium access control, TDMA, underwater acoustic sensor networks.

This manuscript was submitted to IEEE Access on 12 Jun 2020.<div><br></div><div>Abstract:</div><div><br></div><div>Simulation forms an important part of the development and empirical evaluation of underwater acoustic network (UAN) protocols. The key feature of a credible network simulation model is a realistic channel model. A common approach to simulating realistic underwater acoustic (UWA) channels is by using specialised beam tracing software such as BELLHOP. However, BELLHOP and similar modeling software typically require knowledge of ocean acoustics and a substantial programming effort from UAN protocol designers to integrate it into their research. In this paper, we bridge the gap between low level channel modeling via beam tracing and automated channel modeling, e.g. via the World Ocean Simulation System (WOSS), by providing a distilled UWA channel modeling tutorial from the network protocol design point of view. The tutorial is accompanied by our MATLAB simulation code that interfaces with BELLHOP to produce channel data for UAN simulations. As part of the tutorial, we describe two methods of incorporating such channel data into network simulations, including a case study for each of them: 1) directly importing the data as a look-up table, 2) using the data to create a statistical channel model. The primary aim of this paper is to provide a useful learning resource and modeling tool for UAN protocol researchers. Initial insights into the UAN protocol design and performance provided by the statistical channel modeling approach presented in this paper demonstrate its potential as a powerful modeling tool for future UAN research.<br></div>

No abstract