Application of Linear Algorithms to Localization of Underwater Acoustic Communication

摘要

In this thesis we mainly applies LPP (locality preserving projection) MFA (marginal fisher analysis) and ILPP (iterative locality preserving projection) are applied localization of underwater acoustic communication In this study to reduce the impact of multi-path reflection on measured signals localization of underwater communication is based on fingerprinting approaches The procedure is the same as human fingerprint identification The process of localization is divided into two parts including gathering signal in the off-line (i e training) stage and estimating the current localization in the on-line (i e testing) stage In the off-line stage we received intensity of underwater acoustic signals stored in the database at different reference location Then the database is processed through three linear algorithms including LPP MFA and ILPP The purpose of three algorithms is to reduce the computation complexity and economize time to gather the signals in advance to increase the efficiency of localization In the on-line stage we used ML (maximum likelihood) to estimate coordinate of underwater acoustic signals And positioning error is calculate distance between actual position and estimate position by euclidean distance in underwater environment The experiment results show three linear algorithms can achieve high efficiency of localization and it’s possible to realize localization of underwater communication This thesis is divided into six chapters The first chapter is Introduction The second chapter is an explanation of algorithmic theories including LPP MFA and ILPP The third fourth and fifth chapter apply LPP MFA and ILPP to localization of underwater communication The sixth chapter is conclusion of this thesis and suggested direction of future research

獎項日期	2014 7月 31
原文	English
監督員	Kun-Chou Lee (Supervisor)