Building a database of ocean channel impulse responses for underwater acoustic communication performance evaluation: issues, requirements, methods and results
{"title":"Building a database of ocean channel impulse responses for underwater acoustic communication performance evaluation: issues, requirements, methods and results","authors":"T. C. Yang, S. H. Huang","doi":"10.1145/2999504.3001082","DOIUrl":null,"url":null,"abstract":"With a growing number of modulation and equalization schemes proposed for underwater acoustic communications, the need exists for methods to evaluate the performance of the various methods under the same channel conditions. An ideal solution is to develop channel simulators using acoustic propagation models. However, this effort has not been fruitful since acoustic channel simulations so far have failed to capture the characteristics of real ocean channels at practical communication frequencies, except for special cases where the acoustic environments are well known. Currently, acoustic communications still rely on field experiments to evaluate their performance. The problem with the experimental approach is that each experiment is usually tailored to a particular modulation scheme and the cost usually prohibits testing of many different algorithms under the same ocean conditions. Some data-based channel simulators have been proposed where realizations of the channel impulse response (CIR) functions are generated based on the scattering functions estimated from real data but this method is so far not adequate. A scheme to build ocean channel database is proposed in this paper, where CIR is deduced from data on the time scale of a fractional (half) symbol to capture the fine time scale changes of the channel. The estimated CIRs are good replica of the true CIRs as evidenced by the negligibly small ( < −20 dB) channel estimation error or the surrogate signal predication error. The CIR database will allow the developers to test the performance of different algorithms under the same realistic ocean conditions without having to conduct experiments at sea.","PeriodicalId":378624,"journal":{"name":"Proceedings of the 11th International Conference on Underwater Networks & Systems","volume":"4 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 11th International Conference on Underwater Networks & Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2999504.3001082","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
With a growing number of modulation and equalization schemes proposed for underwater acoustic communications, the need exists for methods to evaluate the performance of the various methods under the same channel conditions. An ideal solution is to develop channel simulators using acoustic propagation models. However, this effort has not been fruitful since acoustic channel simulations so far have failed to capture the characteristics of real ocean channels at practical communication frequencies, except for special cases where the acoustic environments are well known. Currently, acoustic communications still rely on field experiments to evaluate their performance. The problem with the experimental approach is that each experiment is usually tailored to a particular modulation scheme and the cost usually prohibits testing of many different algorithms under the same ocean conditions. Some data-based channel simulators have been proposed where realizations of the channel impulse response (CIR) functions are generated based on the scattering functions estimated from real data but this method is so far not adequate. A scheme to build ocean channel database is proposed in this paper, where CIR is deduced from data on the time scale of a fractional (half) symbol to capture the fine time scale changes of the channel. The estimated CIRs are good replica of the true CIRs as evidenced by the negligibly small ( < −20 dB) channel estimation error or the surrogate signal predication error. The CIR database will allow the developers to test the performance of different algorithms under the same realistic ocean conditions without having to conduct experiments at sea.