A fundamental study to develop a wireless sensor network based home system for elderly people and people who are sick was developed by Anuroop Gaddam during his work as PhD student at Massey University in New Zealand supervised by Dr. Subhas Chandra Mukhopadhyay, a Professor of Sensing Technology at the School of Engineering and Advanced Technology. The wireless elderly monitoring system demonstrates that smart, simple sensor devices can be used to recognize activities of daily living and life style of elderly person living alone. The system is becoming more complete as more intelligent features are added to detect the daily activity patterns efficiently. Even though it uses a limited number of sensors, he explains that the system can capture the abnormality in a person’s daily routine by recognizing the use of appliances necessary for daily living, thus determining the life style of elderly person living alone.
The system can be installed/maintained in residential environments without any complexity. Moreover, the proposed sensing system presents an alternative to sensors that are perceived by most people as invasive such as use of cameras and microphones, making the sensors are almost invisible to the user thereby increasing the acceptance level to use the system in a house hold environment.
According to the inventor, the system can be easily installed in an existing home environment with no major modifications or damage. The developed system continuously monitor the activity of the elderly person staying alone and generate the sensor activity pattern to analyse and foresee the changes in daily activities of the elderly person. In the near future, the generated activity pattern related to weekdays and weekends will be used to predict the unusual behaviour of the elderly person based on the classification model of regular and irregular sensor activity.
The results of the Real-time activity behaviour recognition of the inhabitant he obtained were encouraging. Classifications of abnormal situation based on multi-level check method were also validated with machine learning methods. Integration of wellness function in the developed software system was effectively able to determine the abnormal situation of the inhabitant.
In a brief introduction, he explained that statistics shows that there is increasing number of elderly people around the world and this is not going to change. At old age, elderly people are not physically strong enough and are prone to different types of accidents, creating the need to increase hospital care. In some situation they are not allowed to be at their own home. If they prefer to live alone they do however require constant monitoring so that help can be provided immediately in times of dire needs. The best solution is a wireless solution as it allows the units to be small, easy to install, and much more convenient. By integrating sensing in the house hold devices would allow people to be monitored constantly. The benefit by this approach is to help increase health monitoring of elderly people.
Moreover, Anuroop reviewed the state of the art research in the field of the smart home monitoring. Various issues, methodologies and methods involved in development in this field. Modern smart sensors, together with advanced wireless communication technology, can assist in building systems that can be deployed to monitor people non-invasively in real time. Such systems are cost effective, easy to install and maintain and provide a great sense of security to not only the person living alone but also to the family and care givers. It is envisaged that with the rapid increase of the population of the aged people in the world, such systems will find wide acceptance and become prevalent.
He detailed several of the sensors which can be used effectively to build a smart home monitoring system. The design and various issues involved in developing and integrating the sensors. More specifically reviews on Smart homes using audio – visual based systems, wearable sensors and sensors for tracking/monitoring various appliances in a home.
Anuroop explained the design and implementation of the first prototype of the wireless sensor network based home monitoring system. This wireless monitoring system is called “Selective Activity Monitoring (SAM) system”. SAM is an electronic system designed to support people who wish to live alone but, because of age, a health problem or disability, there is some risk in this which worries their family or friends. The system offers such people an unobtrusive safety net which monitors the activity of appliances throughout the house and contacts family members or close friends upon unusual activity. The SAM system is capable of wirelessly sending a text message to a specified cell phone number, this allows the system to send for help, should help be required. This system is a proof of concept that has been successfully designed and fabricated.
The stipulated system was seconded by an intelligent home monitoring unit based on ZigBee wireless sensors has been designed and developed to assist and monitor the inhabitant living in smart environment. The system works on the principle of using sensor units to monitor the basic household appliances used for day-to-day life of the elderly person being monitored. A central coordinator of the sensors collects the data from the sensors connected to various appliances.
Electrical appliance monitoring unit had been modified to monitor more than one appliance to reduce the sensor-count. The electrical appliance monitoring unit is fabricated to accommodate three different electrical appliances on a single power inlet, having the intelligence to detect which particular device is ON and for how long it is used. The new version of bed monitoring sensor incorporated with monitoring system to detect the abnormalities in the person’s daily sleep window period. A temperature-humidity sensor unit and contact sensor has been added to the system to detect the inhabitants patter more effectively. The panic button is modified to send not only emergency message bit also to deactivate any false alarms raised.
Finally, Anuroop describes the wireless sensors setup, communication and an intelligent process for person’s activity behaviour detection. Integrated wireless sensors unit are fabricated to function properly with the house-hold appliances in turn are used for identifying and recognizing the habitual nature of the elderly person. Based on a survey among inhabitants he found out that it has a huge acceptability to be used at home.
The intelligent software, along with the electronic system, can monitor the usage of different household electronic appliances and recognize the activity pattern in real-time. Also, the system interprets all the essential elderly activities such as preparing breakfast/lunch/dinner, showering, rest room use, dinning, sleeping and self-grooming.
Basically, the system functions on the usage of electrical and non-electrical appliances within a home. At hardware level, wireless sensor network with ZigBee components are connected in form of star topology and a central coordinator of the sensors collect the data from the sensors connected to various appliances of the smart home. The developed intelligent program continuously reads the data from the coordinator and efficiently stores on the system for further data processing in real time.
The data processing involves steps for check on the knowledge base for determining the normal activity pattern of the inhabitant. In this system, required numbers of sensors for monitoring the basic activity of inhabitant are used as this is sufficient for monitoring essential activities of elderly behaviour.
The work conducted by Anuroop was focused on the development of a wireless sensors network based smart home monitoring system for elder-care. As a result a wireless sensors network based smart home monitoring system for elder-care was successfully developed to deploy in a real-life scenario. Further, more areas have also opened for investigation in the future. In his conclusion, several areas are suggested here for future research on the topic.
1. The nature of the system lends itself to being expanded further to create a more versatile system. As the wireless sensor network is scalable, there is need to include sensors related to remaining house-hold appliances at elderly house in order to study highly structured behavioural pattern for more precise abnormal detection.
2. Extensive analyses need to be conducted in predicting the abnormal situation of the inhabitant for the developed system.
3. The sensor should be reduced in size drastically by using only ZigBee modules. It has been noticed that the need of a microcontroller can be eliminated by using the ZigBee chip efficiently. The ideal design of the monitoring system would hide the sensor components from the inhabitant almost completely
4. The sensors need to be designed with more precision, enabling to be installed in a house-hold easily.
Source