For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Human Motion Classification Using Radio Signal Strength in WBAN
Sukhumarn ARCHASANTISUK Takahiro AOYAGI Tero UUSITUPA Minseok KIM Jun-ichi TAKADA
IEICE TRANSACTIONS on Communications
Publication Date: 2016/03/01
Online ISSN: 1745-1345
Type of Manuscript: Special Section PAPER (Special Section on Information and Communication Technology for Healthcare and Medical Applications in Conjunction with Main Topics of ISMICT2015)
human motions, classification, radio signal strength, WBAN, simulation,
Full Text: PDF>>
In this paper, a novel approach of a human motion classification system in wireless body area network (WBAN) using received radio signal strength was developed. This method enables us to classify human motions in WBAN using only the radio signal strength during communication without additional tools such as an accelerometer. The proposed human motion classification system has a potential to be used for improving communication quality in WBAN as well as recording daily-life activities for self-awareness tool. To construct the classification system, a numerical simulation was used to generate WBAN propagation channel in various motions at frequency band of 403.5MHz and 2.45GHz. In the classification system, a feature vector representing a characteristic of human motions was computed from time-series received signal levels. The proposed human motion classification using the radio signal strength based on WBAN simulation can classify 3-5 human motions with the accuracy rate of 63.8-95.7 percent, and it can classify the human motions regardless of frequency band. In order to confirm that the human motion classification using radio signal strength can be used in practice, the applicability of the classification system was evaluated by WBAN measurement data.