Open AccessActivity recognition (AR) to detect dementia using wi-fi based wireless sensor network with receive signal srength indicator (RSSI) method
Author(s) -
Hani Rubiani,
E. Samsoleh,
Sulidar Fitri
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1115/1/012079
Subject(s) - dementia , wireless sensor network , computer science , activity recognition , signal (programming language) , life expectancy , population , wireless , pattern recognition (psychology) , real time computing , speech recognition , artificial intelligence , medicine , telecommunications , disease , computer network , environmental health , pathology , programming language
The existence of an increase in human life is the main goal of the growing development of information technology. There are several technologies in the field of Wireless Sensor Network (WSN) which are the main features, one of which is activity recognition. Using AR allows understanding human behavior when carrying out an activity. We have entered a period of population aging, namely an increase in life expectancy which increases with increasing old age. The elderly is susceptible to a disease called dementia. One of the symptoms of dementia is repetition of simple activities such as wandering. These symptoms can be detected by the AR system by classifying the current activity patterns of the elderly and comparing them with previously stored patterns to show early symptoms such as wandering. In efficient AR location research and Wi-Fi based detection methods using the Receive Signal Strength Indicator (RSSI) method and location detection algorithms with K-NN modeling with k = 1. With cross validation showing k = 1 yields an accuracy of 75.73% and distance The minimum average error is 0.298 meters which is better than k = 2 to k = 10.To see the pattern of object movement, in this case, the 85 year old elderly applies the Random Way Point (RWP) model as a model to infer or conclude dementia with calculating the value of the A total based on the location of the object and the time it was observed for 24 hours, and the result is that the amount of 3.139 exceeds the limit of normal conditions of 1.81.