2019Unpublished venueRequires access

Multipath Error Reduction for Underwater Acoustic Direction of Arrival Estimation

Keisaku Takada, Shingo Yoshizawa, Tomoya Tsukui, Takashi Saito

Open publisher page 4 citations

Abstract

This paper presents underwater acoustic direction of arrival (DOA) estimation in highly reflective multipath environments. In ultra-short baseline (USBL), DOA is estimated from a time difference of arrival (TDOA). TDOA is detected by the peak position of cross-correlation values between two received signals. Strong reverberation causes outliers in DOA estimation. We focus on improvement of peak detection to reduce their outliers and propose a new method of IR-GCC-PHAT. The proposed method has been evaluated in both simulation and experiment and shown fewer outliers than the other methods.

About this research paper

What this paper is about

This paper presents underwater acoustic direction of arrival (DOA) estimation in highly reflective multipath environments. In ultra-short baseline (USBL), DOA is estimated from a time difference of arrival (TDOA). TDOA is detected by the peak position of cross-correlation values between two received signals. Strong reverberation causes outliers in DOA estimation. We focus on improvement of peak detection to reduce their outliers and propose a new method of IR-GCC-PHAT. The proposed method has been evaluated in both simulation and experiment and shown fewer outliers than the other methods.

Why it matters

OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper presents underwater acoustic direction of arrival (DOA) estimation in highly reflective multipath environments. In ultra-short baseline (USBL), DOA is estimated from a time difference of arrival (TDOA). TDOA is detected by the peak position of cross-correlation values between two received signals. Strong reverberation causes outliers in DOA estimation. We focus on improvement of peak detection to reduce their outliers and propose a new method of IR-GCC-PHAT. The proposed method has been evaluated in both simulation and experiment and shown fewer outliers than the other methods.

Key concepts: Multilateration, Reverberation, Direction of arrival, Outlier, Multipath propagation, Computer science, Underwater, Arrival time

Related papers

Back to paper searchBrowse research topicsOriginal source
Multipath Error Reduction for Underwater Acoustic Direction of Arrival Estimation — Research Paper | ScholarLens