HUSCAP logo Hokkaido Univ. logo

Hokkaido University Collection of Scholarly and Academic Papers >
Theses >
博士 (情報科学) >

Indoor Acoustic Localization using Reflected Signals

Files in This Item:
Hiroaki_Murakami.pdf10.62 MBPDFView/Open
Please use this identifier to cite or link to this item:https://doi.org/10.14943/doctoral.k14584
Related Items in HUSCAP:

Title: Indoor Acoustic Localization using Reflected Signals
Other Titles: 反射波を用いることによる屋内音響測位に関する研究
Authors: 村上, 弘晃 Browse this author
Issue Date: 25-Mar-2021
Publisher: Hokkaido University
Abstract: Positional information is often required in various situations with the popularization of a smartphone.In particular, COVID-19 epidemicity forces us to control social distance and keep track of contact history with infected people, which means that the movement that require user’s high-accurate positioning information is becoming stronger.The global navigation satellite system (GNSS) is used as a general position-estimation system.However, GNSS signals do not reach indoors, and thus we cannot use GNSS.For the reason, the study on indoor positioning technology is being actively undertaken. In this paper, we assume high-accuracy positioning with an error of less than 0.20m, which can be distinguished an user from neighboring people.Therefore, we focus on acoustic signals that satisfy this requirement and its signals can be received at a smartphone as with GNSS signals.In the positioning using acoustic signals, the modulated signal is transmitted from speakers, and received by the two microphones built in a smartphone.Speakers for the signal transmission are widely used as infrastructure,which means that the user’s position can be estimated without the need for additional equipment.On the other hand, the positioning method using a smartphone requires the arrival time of signals transmitted from at least four speakers in order to estimate the 3-D position. In Japan, the Fire service Act requires speakers to be installed indoors, thus there are speakers indoors absolutely.However, there are few environments with four or more speakers that the user can use, and there are only one or two speakers available in many situations.Therefore, one of the problems to be solved in order to be used acoustic positioning as a standard technology is that the number of speakers required for positioning is insufficient compared to the number of speakers that can be used in indoor facilities. The purpose of this paper is to establish a positioning method in an environment where the number of speakers required for positioning is insufficient.In this paper, we work on several approaches that actively utilize the reflected signal, which is known as a factor of positioning error in acoustic positioning, and achieve this purpose.Two situations is assumed: a corridor where reflected signals from the walls and floor are constantly generated, and a the room where only a reflected signal from the floor are generated and there are obstacles including furniture.Their solutions are shown in this paper. Through evaluation experiments, we confirmed that the user’s position-estimation was possible using only one or two speakers for either proposed method, and highly accurate positioning was achieved by taking advantage of the characteristics of acoustic signals.We also discussed the effectiveness of the proposed methods and its limitations,and mentioned future work.
スマートフォンの普及に伴い,様々な場面において位置情報を求められる動きが強まっている.特に,COVID-19 に伴うソーシャルディスタンスの制御や,感染者との接触歴把握等,ユーザの高精度な位置情報は今後必要不可欠となる.広く使われている位置推定システムにglobal navigation satellite system (GNSS) があるが,屋内ではGNSS 信号が遮蔽され利用できない.そのため、屋内測位技術に関する研究が盛んに取り組まれている。 本論文では,隣り合う人との区別が可能な誤差0.20m 以内での高精度な測位を前提とする.そのため,GPS のようにスマートフォンで受信可能な信号で,この要求を満たす音響信号に着目する.音響信号を用いる測位では,変調された信号がスピーカから送信され,スマートフォンに搭載される2つのマイクロフォンで受信される.音響信号を送信するスピーカは,屋内施設のインフラストラクチャとして既に普及しているため,追加の機材を必要とせずにユーザの位置を推定することができる.一方で,スマートフォンを用いた測位方式では,3次元測位を行うために,最低でも4台のスピーカから送信される信号の到来時刻を必要とする. 日本では消防法により屋内施設でのスピーカ設置義務があり,屋内には必ずスピーカが設置される.しかし,その法律上任意の地点においてユーザが利用可能なスピーカが4台以上ある環境はほとんどなく,1 台もしくは2 台である環境が多い.そのため,音響測位を標準技術として展開するために解決すべき課題の1つに,屋内施設にて利用可能なスピーカ台数と比較し,測位に必要なスピーカの台数が不十分な点が挙げられる. 本論文では,測位に必要なスピーカ台数が不十分な環境における測位手法の確立を目的とする.本稿では,音響測位において測位誤差の要因として知られる反射波をあえて積極的に活用したアプローチに取り組み,この目的を実現する.定常的に壁と床からの反射波が生成される廊下と,床からの反射波のみが生じ,家具などの障害物がある部屋の2つの状況を想定し,それぞれ異なる解決策を示す. 評価実験を通し,いずれの手法に対しても1台もしくは2台のスピーカ台数で測位ができることを確認し,音響信号の特性を活かした高精度な測位を達成した.また,提案手法の有効性とその限界について論じ,今後の課題について言及する.
Conffering University: 北海道大学
Degree Report Number: 甲第14584号
Degree Level: 博士
Degree Discipline: 情報科学
Examination Committee Members: (主査) 教授 杉本 雅則, 教授 田中 章, 教授 工藤 峰一, 教授 今井 英幸
Degree Affiliation: 情報科学研究科(情報理工学専攻)
Type: theses (doctoral)
URI: http://hdl.handle.net/2115/81248
Appears in Collections:課程博士 (Doctorate by way of Advanced Course) > 情報科学院(Graduate School of Information Science and Technology)
学位論文 (Theses) > 博士 (情報科学)

Export metadata:

OAI-PMH ( junii2 , jpcoar_1.0 )

MathJax is now OFF:


 

 - Hokkaido University