| Presentation | 2010-07-16 State Classification based on Signal Subspace using Support Vector Machine for Wireless Monitoring Jihoon HONG, Shun KAWAKAMI, Tomoaki OHTSUKI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this report, we propose an indoor-outdoor state classification system based on signal subspace by eigenvector, which uses Support Vector Machines(SVMs)and can monitor states in a room or inside and around a vehicle, by using only one transmitter and one array antenna receiver. Electric waves sent from a transmitter reflect, diffract and penetrate throughout the environment. The characteristics of the received signal can represent several propagation environments, such as that of a whole room(indoor)or inside and around a car(outdoor). By using array signal processing, we get an eigenvector spanning signal subspace inherent to each environment, which removes the influences of fading and noise. We then extract features for each state from the eigenvector data, and classify each state using SVMs. We performed experiments to classify seven states in an indoor setting: "No event," "Walking into the bathroom," "Entering into the bathtub," "Standing while showering," "Sitting while showering," "Falling down", and "Passing out;" and two states in an outdoor setting: "Normal situation" and "Abnormal situation." Experimental results showed that the proposed method can correctly classify all the states, with 100% accuracy. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | state classification / monitoring / array antenna / Support Vector Machine |
| Paper # | RCS2010-72 |
| Date of Issue |
| Conference Information | |
| Committee | RCS |
|---|---|
| Conference Date | 2010/7/8(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Radio Communication Systems (RCS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | State Classification based on Signal Subspace using Support Vector Machine for Wireless Monitoring |
| Sub Title (in English) | |
| Keyword(1) | state classification |
| Keyword(2) | monitoring |
| Keyword(3) | array antenna |
| Keyword(4) | Support Vector Machine |
| 1st Author's Name | Jihoon HONG |
| 1st Author's Affiliation | Graduate School of Science and Technology, Keio University() |
| 2nd Author's Name | Shun KAWAKAMI |
| 2nd Author's Affiliation | Graduate School of Science and Technology, Keio University |
| 3rd Author's Name | Tomoaki OHTSUKI |
| 3rd Author's Affiliation | Department of Information and Computer science, Keio University |
| Date | 2010-07-16 |
| Paper # | RCS2010-72 |
| Volume (vol) | vol.110 |
| Number (no) | 127 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |