Abstract
Determining the relationship between the performance of an underwater acoustic data communications system and the operating environmental conditions is a problem that continues to plague researchers. The complexity of the time-varying channel is difficult to measure and model. Therefore an approach that uses metrics measured from data collected at sea to characterize the channel is attractive. As expected, preliminary assessments on limited data have shown that performance depends not only on environmental conditions, but also on system implementation. By extracting a variety of metrics, a better understanding of the subset that discriminate between good and bad performance can be developed. Also by analyzing the relationship between certain metrics and performance, system limitations can be identified for re-evaluation. For example, a surprising result of the initial assessment of performance using a multichannel decision feedback equalizer on real data showed that sparseness of multipath arrivals may be an arbiter of performance [Richman et al., J. Acoust. Soc. Am. 110, 2619 (2001)]. Therefore changes to the algorithm that allows for sparse arrivals may improve performance. In this paper, a larger number of metrics from greater quantities of real-data and system configurations are measured and evaluated against equalizer results.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call