Indoor Event Detection with Eigenvector Spanning Signal Subspace for Home or Office Security

Shohei IKEDA; Hiroyuki TSUJI; Tomoaki OHTSUKI

doi:10.1587/transcom.E92.B.2406

Indoor Event Detection with Eigenvector Spanning Signal Subspace for Home or Office Security

Shohei IKEDA, Hiroyuki TSUJI, Tomoaki OHTSUKI

Full Text Views

0

Cite this

Summary :

This paper proposes an indoor event detection system for homes and offices that is based on electric wave reception such as intrusion into home or office. The proposed system places antenna array on the receiver side and detects events such as intrusion using the eigenvector spanning signal subspace obtained by the antenna array. The eigenvector is based on not received signal strengths (RSS) but direction of arrival (DOA) of incident signals on the antenna array. Therefore, in a static state, the variance of the eigenvector over time is smaller than that of RSS. The eigenvector changes only when the indoor environment of interest changes intermittently and statically, or dynamically. The installation cost is low, because the detection range is wide owing to indoor reflections and diffraction of electric wave and only a pair of transmitter and receiver are used. Experimental results reveal that the proposed method can distinguish the state when no event occurs and that when an event occurs clearly. Since the proposed method has a low false detection rate, it offers higher detection rates than the systems based on RSS.

Publication: IEICE TRANSACTIONS on Communications Vol.E92-B No.7 pp.2406-2412

Publication Date: 2009/07/01

Publicized

Online ISSN: 1745-1345

DOI: 10.1587/transcom.E92.B.2406

Type of Manuscript: PAPER

Category: Wireless Communication Technologies

Cite this

Copy

Shohei IKEDA, Hiroyuki TSUJI, Tomoaki OHTSUKI, "Indoor Event Detection with Eigenvector Spanning Signal Subspace for Home or Office Security" in IEICE TRANSACTIONS on Communications, vol. E92-B, no. 7, pp. 2406-2412, July 2009, doi: 10.1587/transcom.E92.B.2406.
Abstract: This paper proposes an indoor event detection system for homes and offices that is based on electric wave reception such as intrusion into home or office. The proposed system places antenna array on the receiver side and detects events such as intrusion using the eigenvector spanning signal subspace obtained by the antenna array. The eigenvector is based on not received signal strengths (RSS) but direction of arrival (DOA) of incident signals on the antenna array. Therefore, in a static state, the variance of the eigenvector over time is smaller than that of RSS. The eigenvector changes only when the indoor environment of interest changes intermittently and statically, or dynamically. The installation cost is low, because the detection range is wide owing to indoor reflections and diffraction of electric wave and only a pair of transmitter and receiver are used. Experimental results reveal that the proposed method can distinguish the state when no event occurs and that when an event occurs clearly. Since the proposed method has a low false detection rate, it offers higher detection rates than the systems based on RSS.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E92.B.2406/_p

Copy

@ARTICLE{e92-b_7_2406,
author={Shohei IKEDA, Hiroyuki TSUJI, Tomoaki OHTSUKI, },
journal={IEICE TRANSACTIONS on Communications},
title={Indoor Event Detection with Eigenvector Spanning Signal Subspace for Home or Office Security},
year={2009},
volume={E92-B},
number={7},
pages={2406-2412},
abstract={This paper proposes an indoor event detection system for homes and offices that is based on electric wave reception such as intrusion into home or office. The proposed system places antenna array on the receiver side and detects events such as intrusion using the eigenvector spanning signal subspace obtained by the antenna array. The eigenvector is based on not received signal strengths (RSS) but direction of arrival (DOA) of incident signals on the antenna array. Therefore, in a static state, the variance of the eigenvector over time is smaller than that of RSS. The eigenvector changes only when the indoor environment of interest changes intermittently and statically, or dynamically. The installation cost is low, because the detection range is wide owing to indoor reflections and diffraction of electric wave and only a pair of transmitter and receiver are used. Experimental results reveal that the proposed method can distinguish the state when no event occurs and that when an event occurs clearly. Since the proposed method has a low false detection rate, it offers higher detection rates than the systems based on RSS.},
keywords={},
doi={10.1587/transcom.E92.B.2406},
ISSN={1745-1345},
month={July},}

Copy

TY - JOUR
TI - Indoor Event Detection with Eigenvector Spanning Signal Subspace for Home or Office Security
T2 - IEICE TRANSACTIONS on Communications
SP - 2406
EP - 2412
AU - Shohei IKEDA
AU - Hiroyuki TSUJI
AU - Tomoaki OHTSUKI
PY - 2009
DO - 10.1587/transcom.E92.B.2406
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E92-B
IS - 7
JA - IEICE TRANSACTIONS on Communications
Y1 - July 2009
AB - This paper proposes an indoor event detection system for homes and offices that is based on electric wave reception such as intrusion into home or office. The proposed system places antenna array on the receiver side and detects events such as intrusion using the eigenvector spanning signal subspace obtained by the antenna array. The eigenvector is based on not received signal strengths (RSS) but direction of arrival (DOA) of incident signals on the antenna array. Therefore, in a static state, the variance of the eigenvector over time is smaller than that of RSS. The eigenvector changes only when the indoor environment of interest changes intermittently and statically, or dynamically. The installation cost is low, because the detection range is wide owing to indoor reflections and diffraction of electric wave and only a pair of transmitter and receiver are used. Experimental results reveal that the proposed method can distinguish the state when no event occurs and that when an event occurs clearly. Since the proposed method has a low false detection rate, it offers higher detection rates than the systems based on RSS.
ER -