Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications

Mojtaba Farmani, Michael Syskind Pedersen, Zheng-Hua Tan, Jesper Jensen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

7 Citations (Scopus)

Abstract

This paper deals with estimation of the target sound Direction of Arrival (DoA) or a Hearing Aid System (HAS) which can connect to a wireless microphone worn by target talker. In this setup, the HAS is "informed" about the almost noise-free content of the target sound via the wireless microphone and can use this information for the DoA estimation. Here, we propose an "informed'' DoA estimator based on the Time Difference of Arrival (TDoA) of the target sound at two microphones mounted on the ears of the HAS user---one microphone on each ear. To estimate the TDoA and the DoA, we propose a maximum likelihood framework relying on the noise-free target sound and estimation of the ambient noise characteristics. We show how the proposed ML framework allows us to estimate the TDoA and the DoA jointly or consecutively. Further, to evaluate the likelihood function efficiently, we resort to an Inverse Discrete Fourier Transform (IDFT) technique. To study the performance of the proposed algorithms, we run simulations for various DoAs, Signal to Noise Ratios (SNRs), and distances in large crowd noise situations. In these situations, the proposed estimator improves the estimation performance markedly over a recently proposed "informed'' TDoA-based DoA estimator.
Original languageEnglish
Title of host publication2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP)
PublisherIEEE Signal Processing Society
Publication date14 Dec 2015
Pages953 - 957
ISBN (Electronic)978-1-4799-7591-4
DOIs
Publication statusPublished - 14 Dec 2015
Event3rd IEEE Global Conference on Signal and Information Processing (GlobalSIP'2015) - Orlando, FL, United States
Duration: 14 Dec 201516 Dec 2015

Conference

Conference3rd IEEE Global Conference on Signal and Information Processing (GlobalSIP'2015)
CountryUnited States
CityOrlando, FL
Period14/12/201516/12/2015
SeriesIEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings

Fingerprint

Hearing aids
Direction of arrival
Microphones
Acoustic noise
Acoustic waves
Information use
Discrete Fourier transforms
Maximum likelihood
Time difference of arrival
Signal to noise ratio

Cite this

Farmani, M., Pedersen, M. S., Tan, Z-H., & Jensen, J. (2015). Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications. In 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP) (pp. 953 - 957). IEEE Signal Processing Society. IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings https://doi.org/10.1109/GlobalSIP.2015.7418338
Farmani, Mojtaba ; Pedersen, Michael Syskind ; Tan, Zheng-Hua ; Jensen, Jesper. / Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications. 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE Signal Processing Society, 2015. pp. 953 - 957 (IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings).
@inproceedings{d1e8174f9af64723a4dc103a5e668045,
title = "Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications",
abstract = "This paper deals with estimation of the target sound Direction of Arrival (DoA) or a Hearing Aid System (HAS) which can connect to a wireless microphone worn by target talker. In this setup, the HAS is {"}informed{"} about the almost noise-free content of the target sound via the wireless microphone and can use this information for the DoA estimation. Here, we propose an {"}informed'' DoA estimator based on the Time Difference of Arrival (TDoA) of the target sound at two microphones mounted on the ears of the HAS user---one microphone on each ear. To estimate the TDoA and the DoA, we propose a maximum likelihood framework relying on the noise-free target sound and estimation of the ambient noise characteristics. We show how the proposed ML framework allows us to estimate the TDoA and the DoA jointly or consecutively. Further, to evaluate the likelihood function efficiently, we resort to an Inverse Discrete Fourier Transform (IDFT) technique. To study the performance of the proposed algorithms, we run simulations for various DoAs, Signal to Noise Ratios (SNRs), and distances in large crowd noise situations. In these situations, the proposed estimator improves the estimation performance markedly over a recently proposed {"}informed'' TDoA-based DoA estimator.",
author = "Mojtaba Farmani and Pedersen, {Michael Syskind} and Zheng-Hua Tan and Jesper Jensen",
year = "2015",
month = "12",
day = "14",
doi = "10.1109/GlobalSIP.2015.7418338",
language = "English",
pages = "953 -- 957",
booktitle = "2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP)",
publisher = "IEEE Signal Processing Society",
address = "United States",

}

Farmani, M, Pedersen, MS, Tan, Z-H & Jensen, J 2015, Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications. in 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE Signal Processing Society, IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings, pp. 953 - 957, 3rd IEEE Global Conference on Signal and Information Processing (GlobalSIP'2015), Orlando, FL, United States, 14/12/2015. https://doi.org/10.1109/GlobalSIP.2015.7418338

Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications. / Farmani, Mojtaba; Pedersen, Michael Syskind; Tan, Zheng-Hua; Jensen, Jesper.

2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE Signal Processing Society, 2015. p. 953 - 957.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications

AU - Farmani, Mojtaba

AU - Pedersen, Michael Syskind

AU - Tan, Zheng-Hua

AU - Jensen, Jesper

PY - 2015/12/14

Y1 - 2015/12/14

N2 - This paper deals with estimation of the target sound Direction of Arrival (DoA) or a Hearing Aid System (HAS) which can connect to a wireless microphone worn by target talker. In this setup, the HAS is "informed" about the almost noise-free content of the target sound via the wireless microphone and can use this information for the DoA estimation. Here, we propose an "informed'' DoA estimator based on the Time Difference of Arrival (TDoA) of the target sound at two microphones mounted on the ears of the HAS user---one microphone on each ear. To estimate the TDoA and the DoA, we propose a maximum likelihood framework relying on the noise-free target sound and estimation of the ambient noise characteristics. We show how the proposed ML framework allows us to estimate the TDoA and the DoA jointly or consecutively. Further, to evaluate the likelihood function efficiently, we resort to an Inverse Discrete Fourier Transform (IDFT) technique. To study the performance of the proposed algorithms, we run simulations for various DoAs, Signal to Noise Ratios (SNRs), and distances in large crowd noise situations. In these situations, the proposed estimator improves the estimation performance markedly over a recently proposed "informed'' TDoA-based DoA estimator.

AB - This paper deals with estimation of the target sound Direction of Arrival (DoA) or a Hearing Aid System (HAS) which can connect to a wireless microphone worn by target talker. In this setup, the HAS is "informed" about the almost noise-free content of the target sound via the wireless microphone and can use this information for the DoA estimation. Here, we propose an "informed'' DoA estimator based on the Time Difference of Arrival (TDoA) of the target sound at two microphones mounted on the ears of the HAS user---one microphone on each ear. To estimate the TDoA and the DoA, we propose a maximum likelihood framework relying on the noise-free target sound and estimation of the ambient noise characteristics. We show how the proposed ML framework allows us to estimate the TDoA and the DoA jointly or consecutively. Further, to evaluate the likelihood function efficiently, we resort to an Inverse Discrete Fourier Transform (IDFT) technique. To study the performance of the proposed algorithms, we run simulations for various DoAs, Signal to Noise Ratios (SNRs), and distances in large crowd noise situations. In these situations, the proposed estimator improves the estimation performance markedly over a recently proposed "informed'' TDoA-based DoA estimator.

U2 - 10.1109/GlobalSIP.2015.7418338

DO - 10.1109/GlobalSIP.2015.7418338

M3 - Article in proceeding

SP - 953

EP - 957

BT - 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP)

PB - IEEE Signal Processing Society

ER -

Farmani M, Pedersen MS, Tan Z-H, Jensen J. Informed TDoA-based Direction of Arrival Estimation for Hearing Aid Applications. In 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE Signal Processing Society. 2015. p. 953 - 957. (IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings). https://doi.org/10.1109/GlobalSIP.2015.7418338