CFDAMA-SRR: A MAC Protocol for Underwater Acoustic Sensor Networks
CFDAMA-SRR: A MAC Protocol for Underwater Acoustic Sensor Networks
复制标题
DOI：
10.1109/access.2019.2915929
复制
发表时间：
2019-01-01
期刊：
影响因子：
3.9
通讯作者：
Zakharov, Yuriy V.
中科院分区：
文献类型：
--
作者：
Gorma, Wael;Mitchell, Paul D.;Zakharov, Yuriy V.
水下声传感器网络是一种使能技术，用于许多应用。声学信道的长传播延迟和有限带宽限制了可实现的端到端延迟、信道利用率和公平性之间的权衡。本文提供了新的见解使用组合的免费/按需分配多址（CFDAMA）计划。CFDAMA可以被分类为自适应TDMA，其中容量通常按需分配。CFDAMA与循环请求（CFDAMA-RRs）示出，以最大限度地减少端到端的延迟和最大限度地提高信道利用率水下。它以最小的开销维持节点之间的公平性，并适应水下信道的变化和随时间变化的业务需求。然而，它的性能在很大程度上取决于网络规模。本文的主要贡献是一个新的计划采用循环请求策略在一个系统的方式（CFDAMA-SRR）。综合事件驱动的河床模拟部署在海床上的网络表明，CFDAMA-SRR优于其底层计划，CFDAMA-RR，特别是当传感器节点广泛分布。考虑到节点的位置，新方案有一个偏见，对长延迟需求分配时隙，以提高性能的CFDAMA-RR。说明性的例子表明，分析和模拟结果之间的良好协议。
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.