A Research About Smart Health Caring Home

This is a research article on smart healthcare and home smart health care solutions that can be used as a reference for understanding and applying smart health devices and solutions to improve your health. The article is based on the author’s master’s thesis at the University of Montpellier in France, and the content has been changed to make it simpler and easier to read.

I.Introduction

I.1 Background

In our society, we are facing a growing population of older people, many of whom live in their own homes. For those who are in dangerous situations, healthy attention is ubiquitous. [1] Aging is accompanied by functional frailties that often cause health problems, loss of autonomy, and harm to family relationships. [2]

Chronic diseases, especially cardiovascular and cerebrovascular diseases, arthritic diseases, Alzheimer’s diseases, which are more common in older age groups, are often an important factor in the decompensation of the physical and psychological homeostasis of the elderly. [3] [4] [5]

Most of them have difficulty walking or losing their independence and taking care of themselves. They also experience mood swings, an uncomfortable body and the enormous cost of treatment problems. [6]However, the number of carers available for frequent home visits is limited. As a result, this phenomenon of continued increase in the senior population will result in a tremendous burden and stress for families and society. And the cost of managing the disease is expensive. [7] There is a need to find solutions to improve living conditions and develop more robust, usable, safe but low-cost health systems to reduce the burden on society. [8]

Biomedicine, information technology, health data, wireless networks and mobile interconnect are transforming traditional medicine from an art into a data-driven science. [9] At the right time, in the right place, with a reasonable price to provide a precise medical treatment service. [10] Research indicates that patients can play an important role in improving health outcomes through personal care based on the use of information technology. [11]

The use of innovative technologies and home automation, which can contribute to independence, is often still empirical and requires. By its nature, a multidisciplinary approach to assess their impacts from an economic, social, medical and not just technical point of view. [12]

Most chronic diseases can be diagnosed in the hospital by the doctor, but after the prescription, the treatment in the hospital is very limited. [13] Most surveys and examples show that more and more seniors and patients with chronic conditions want to stay home as long as possible for their health care. [14] [15] [16] The use of everyday technologies (activity detectors, light pathways, robotic assistance, wireless network …) could significantly improve the lives of older people and help them to stay at home in complete safety in addition to a multi-disciplinary action of the social care sector and family carers. [17]

In many countries, the costs of medical care for chronic diseases make up the majority of direct medical expenses. The aging of the population in developed countries will result in higher medical costs in the future. [18] If there is no credible solution for changing the treatment of chronic diseases, it will not be possible to solve the global problem of high medical expenses and maintenance of health care for chronic patients.

However, no scientific evaluation is currently available regarding the interest in improving or maintaining the physical and mental health of these chronically ill patients with the combination of innovation technologies and socio-medical impact. but also socio-economic for the future direction of sustainable economic development.

How to improve people’s health in the face of current and future challenges of aging. Meeting these challenges would lead to an overall reduction of costs for society. How to respond and what solutions to put in place? What is the place of information technologies for these solutions?

When researching health and home care using innovative technologies for the elderly, several types of application domains are concerned. One of these areas is home monitoring where the primary goal is to ensure the safety of older people in the time they are alone in their homes. [19]

Other proposals are specific to a given part of the house such as the bedroom or the bathroom. [20] Other applications focus on monitoring the health status of individuals. [21]

Other areas There are a number of compact portable sensors used for the detection of emergency situations, such as sensors for the observation of vital signs. Several research projects in this area are focused on identification and evaluation. risks associated with walking in everyday life. Some of them aim to characterize the overall cognitive level of seniors who have fallen or are at risk of falling, while another aims to assess the impact of drugs on the behavior of older people. [22]

The increasing adoption of technology, the rise of distance care and and telemedicine, a growing number of intelligent life services are available for the mass market. Connected objects and medical devices measure, analyze, store and sometimes share our health data: this is called e-health. [23]

These are trends that are transforming aging care and the health industry in the process. [24] So we face the challenges. The identification of the different challenges of home care for health and comfort has made it possible to define the main issues: Prevent dependence, reduce hospitalization and treatment costs and improve the quality of life and productivity of services. [25]

I.2 Objective and specifications

The development of “silver economy” and “E-health” is fast. And e-health could be the next industrial revolution.

The three major problems of the elderly have shown us: chronic diseases, difficulty in mobility and the enormous cost of health care. [26] [27] [28] The main objective of the research on this topic is to discover the possibility of creating a new disruptive model or structure to maintain health at home with the innovation technologies and the point of socio-economic perspective [29] [30] [31]

The general objectives are:1) Find the method of innovation to fight the rupture of the autonomy, the mental sensory impairments and the cost of treatment in the elderly.

2) Better management and prevention of chronic disease related to the elderly, innovation technologies but also coordinating the family, health, medico-social and medico-economic actors.

3) Evaluate these new tools and also evaluate the improvement of the quality of life of these elderly people.

With the advancement of innovation, more and more traditional businesses in the field of home care are thinking of digital development strategies in personal services (digital mediation, new uses, infrastructure, equipment, services …).

For my part, my main interest is to adopt new scientific knowledge on e-health and apply it in the near future for the benefit of my family and vulnerable groups in my country.I would like to pursue a career that involves being aware ethics, welfare benefits, care management, intelligence services and management of home environments adapted to aging and people who lose autonomy.

It is in this environment that my tutor, Mr. David LE NORCY, the director of innovation at Synergies @ Venir, offered me this internship.

As mentioned earlier, most research projects in this area are focused on assessing the risks associated with walking around in everyday life. One of these aims to characterize the cognitive level of elderly people who have fallen or are at risk of falling, while another focuses on assessing the impact of medication on behavior. are focused on the networks.The other are interested in databases.But few studies discuss the synergy and coordination of the internet of things, personal assistant, sensors, home automation networks and e-health.

So my study work is based on the synthesis of different innovative technologies in the home by considering the health and social economy factor and referencing some smart home maintenance projects.

My internship work took place in four parts. The first is to study the feasibility of different technologies for the state of the art of home communication solutions capable of building a “smart health” ecosystem for home care and healthcare, then it is to design and propose a “mockup” of an innovative solution for home support; the third is the implementation and realization of this “mockup”, and finally, it is to analyze and synthesize my work.This report therefore also belongs to the fourth part.

I.3 Initial vision

In a home, sensors and actuators are connected via a Wireless Body Network or Wireless Sensor Network.

In order to obtain automatic, continuous and real-time measurement of physiological signals in a wireless body sensor network. Several sensors can be placed on clothing or directly on the body, or implanted in a tissue, which can facilitate the measurement of blood pressure, blood glucose, electroencephalogram, electrocardiogram, electromyogram and heart rate.

The core Body Sensor Network node collects all physiological data, performs limited data processing and functions as the gateway to the wireless sensor network. The actuators operate on the basis of feedback from the occupants or the central computer system. The central computer system collects environmental, physiological and activity data through the network, analyzes them and can send a feedback back to the user or activate the actuators to control devices such as humidifier, light and heat. It also functions as the central home gateway, which sends measured data to health staff / internet service providers.

As needed, users can also actively control their home devices through voice, gestures, mobile phones and other methods to get help to improve the quality of life, health and well-being.

We can also imagine some specific applications.For example, the cell phone of a diabetic patient in real time health monitoring can detect glucose and send it to a doctor for analysis; myocardial infarction can be reduced by monitoring episodic events and other abnormal conditions through the Wireless Body Network; Connected to an Internet medical facility and integrated into a telemedicine system, to monitor health conditions thereby reducing dependency; patients on clinical surveillance.

II. Presentation the internship company

Since 2007, the SYNERGIES @ VENIR group specializes in services to individuals and territories. A dynamic enterprise in full development, SYNERGIES @ VENIR puts economic and social efficiency at the service of the general interest. It develops a range of innovative services in the field of personal services. Today, the group counts:

• 7 human service agencies, AIDE @ VENIR [156], specialized in home care for the elderly and people with disabilities

• 2 seniors services residences, LA ROSERAIE

• 2 intergenerational micro-nurseries, SMALL CHILDREN

• 1 training center, AGIREA

• 1 home-delivery service, POPOTE MINUTE

• 1 third innovative place, A LAB’BONHEUR.SYNERGIES @ VENIR places the human at the heart of its actions and is divided into 3 axes:

• the journey of users

• the career of employees

• the implementation of projects in response to the needs of the territories 

SYNERGIES @ VENIR is composed of more than 400 employees who adhere to the values ​​and the project of the company. SYNERGIES @ VENIR brings together La Roseraie service centers and AIDE @ VENIR service agencies since June 2015.The group develops every year, projects with the objective of being able to meet the demand for services to people in Gironde.With nearly 400 employees and 1000 users, the group is positioned as the 7th employer of Bordeaux Métrople. SYNERGIES @ VENIR is a player in the Silver Economy and the social and solidarity economy.

III. Technology Studies

III.1 Home Care and eHealth Case Studies

Smart home support is a concept that describes a living environment with a digital environment such as sensors, smart devices, networks and natural interactions with humans. It is able to provide residents with practical and adaptive services. . It can often be used to support people with cognitive impairment who live independently.

The smart home, service robots and e-health are provided to assist people in their well-being or as part of the daily assistance to the elderly. Services are used to provide assistance in a variety of everyday situations, such as collecting and transporting objects from the ground, performing cleaning tasks and providing emergency support.

There have been a number of projects in the world for smart home support and e-health. I have done some studies on different projects related to home care with connected objects, home automation and e-health.

Within the TECNALIA Health Technology Unit [32], there are many projects related to e-health. TECNALIA is the largest private research, development and innovation group in Spain and one of the largest in Europe after a merger process of eight technology centers located in the Basque Country (Spain). It involves older people, people with cognitive or physical disabilities, their relatives and caregivers, clinical experts and medical professionals as potential users in the development of the project.

The GERHOME project [33] was carried out by the CSTB and the NICE CHU to secure home care for the elderly by anticipating home accidents or health problems.The Geri @ TIC [34] project is supported by various Alsatian actors including MEDETIC and CNR-Santé for the creation of secure living units for seniors through home automation and telemedicine. The LENA project [35] is initiated by CENTICH and Clair Soleil Saumur with Schneider Electric, a transitional housing program for seniors following a hospital stay.

QoLT (Quality of Life Technology) Center [36] is founded by the NSF (Engineering Science Center) ERC (National Science Foundation) in Pittsburgh. It focuses on the development of intelligent systems and assistive technologies that enable older people and people with disabilities to live more independently.

The Microsoft Research EasyLiving [37] project has developed an architecture and technologies for intelligent environments. The EasyLiving system has evolved with an intelligent user interface, dynamic device configuration, remote control and activity tracking, allowing for flexible interaction between the user and a wide range of tasks and modalities.

The Ambient Intelligence Research Laboratory [38] at Stanford University focuses on research to develop techniques and applications of ambient intelligence in smart homes and offices. intelligent buildings sensitive to occupations. A network of sensing devices is used to monitor workers’ work habits and social interactions, and adaptive personal recommendations are provided to promote ergonomic health and social engagement in the smart home environment.

The CASAS project [39] Smart Home is a multidisciplinary research project at Washington State University. It focuses on creating a smart home environment. In this project, the smart home environment uses smart agents, where residents’ standing and their physical environment are perceived using sensors and where the environment is used by using controllers to improve comfort and safety. This system is simple and lightweight so that the capabilities of the smart home can be deployed, evaluated and scaled accordingly.

The Aware Home Research Initiative (AHRI) [40] at Georgia Institute of Technology is a remarkable project in the field of home care. The technical components involved in the home include context awareness and ubiquitous detection, individual interaction with the home, and intelligent flooring. The project addresses age-specific applications in three ways: the social connections between older parents and their adult children, which provide peace of mind for family members；reinforce aspects of memory that diminish with age；Manage crisis situations so that appropriate external services are notified.

The ALADIN project (Ambient Lighting Assistance for an Aging Population) [41] proposed a magic lighting system for the elderly. The goal of this project is to develop an adaptive lighting system with intelligent open-loop control, which can adapt to the needs of users in various situations, but which also offers intelligent management of green energy. The dynamic lighting system can be beneficial for eye health, sleep quality, mood, cognitive performance, and even the metabolic system of users, particularly for people with chronic and visual impairments. This system allows citizens with mobility or other disabilities to operate environmental systems and devices directly without physically moving to the location of the actuator. It allows the user to control the devices by looking directly through interaction with the gaze.

RoboCare [42] by ISTC-CNR is a prototype of an integrated home environment named RDE (RoboCare Domestic Environment) with cognitive support to improve the daily lives of elderly people at home.

The iCarer project [43] aims to support informal care for older adults to provide interoperable solutions that offer a holistic support service to care in the cloud. Additional services include a personalized support and training program based on e-learning methods, caregiver support, and care and support services. These services combine to provide a general sense of security and a substantial reduction of stress for the caregiver.

In Asia, the TRON project [44] is an open project on the intelligent living environment and assistive technology. The PAPI project and the U-house project [45] were founded in Taiwan as part of the TRON project. Ubiquitous Home has proposed and implemented contextual services in a real smart environment. The Robotic Room and The Sensing Room at the University of Tokyo is another prototype of a smart home system.

Living Lab Taiwan [46] is a project that promotes open innovation in a real-world environment involving real-world uses, and was launched in June 2008. It is operated by Innovative DigiTech-Enabled Applications & Services Institute (IDEAS) Currently, they have developed E-health for certain areas of interest: Respiratory Care System, Personal Health Assistant and Emotional Ring.

III.2 Voice assistant for home support

III.2.1.Panorama of voice assistant for home support

A voice assistant is a software agent that can perform tasks or services for an individual with voice. Sometimes the term “chatbot” is used to refer to virtual assistants generally or specifically those accessed by speech. [47]

Voice assistants like google assistant and Amazon Alexa have been designed to improve our lives by making information and digital services more accessible. [48]

They also ask about their potential role in one of the most important aspects of our lives – health care. Can these voice assistants at least help us heal and stay in shape? And what opportunities do they offer to health professionals who are increasingly motivated to improve home health care? [49]

Interaction methodVoice assistants work via Voice, for example with Amazon Alexa on the Amazon Echo or Siri device on an iPhone.

Some wizards are available through several methods, such as Google Assistant via Google Allo app chat and Google Home smart speaker voice.

Virtual Assistants use natural language processing to match user text or voice input with executable commands. Many are learning continuously using artificial intelligence techniques, including machine learning.

To activate a wizard using the voice, a wakeup word can be used. This is a word or groups of words such as “Alexa”, “OK Google” or “Hey Siri”.

Voice assistance devices for home support

Today, our homes and appliances are smarter and more connected than ever.

Amazon released Alexa in 2014, and did so by packing it into an Internet-connected home appliance, the Amazon Echo. This device has changed the way people perceive and use voice assistants. Instead of having to talk in a smartphone, the Echo is a device always on / always with a range of sensitive microphones and a built-in speaker. It is a stand-alone voice assistant designed specifically for home use.Google responded to Amazon’s success by announcing their own Amazon Echo equivalent – Google Home. Apple, which has set the bar for voice assistants with Siri, has not yet produced anything similar to Amazon Echo or Google Home, although there are rumors of an upcoming announcement.WHY a voice support for home support?

.AccessibilityThe ability of seniors to view text on a smartphone screen, or on any screen, compromises their ability to take advantage of many digital health care solutions.Voice applications are a natural alternative for these visually impaired people. [50] The more natural interface of voice assistants is also promising for technophobes and older populations where the use of smartphones, tablets and PCs is still lagging behind the general population.

.Ease of useVoice interfaces allow for more natural and conversational interaction.

Engaging with both a hands-free and no-vision interface also means that it can be used in environments where hands or eyes are otherwise occupied, or at home during cooking, cleaning, dressing, bathing, etc. [51]

.SpeedMost people can talk faster than they can type. The average speech rate for an adult is between 150 and 190 words per minute. While the average typing speed is about 40 words per minute. It’s on the side of the entrance. When it comes to response times, we can be ready to wait a few seconds for a website to load. [52] With the power of cloud computing and faster, more ubiquitous network connectivity, the response time or latency of voice assistants has improved.

.Improving accuracyAlthough not yet perfect, voice technology has improved a lot, and even small improvements can have a big impact on user satisfaction. [53]

The proof of these improvements is both anecdotal and empirical. As recently, Google has announced a breakthrough in their voice search technology that has resulted in dramatic improvements in speech recognition.

Microsoft recently announced a breakthrough in their speech recognition technology by demonstrating a precision equivalent to human transcription.

. ExtensibilityMost of the most popular skills for the voice assistant platform involve access and control of third-party connected devices [54] – for example, connected lighting fixtures, thermostats, and audio systems.For example, in its role as a device control center, Amazon Echo becomes both an access point to information and the ultimate universal remote control.Voice assistance for home health.

Voice applications have a long history of clinical care. Companies like Nuance, a leading provider of speech recognition technologies, have long offered solutions that allow clinicians to dictate care notes for transcription and storage.

But while technologies for individuals like google assistant and Alexa have become more common, the role of these new technologies in home care and personal wellness is taken more seriously.

Chronic Disease Management

Patients with a chronic condition such as cardiovascular [55] have a voice-activated assistant who provides information and feedback on self-care, reminders about tests, appointments, and better coordination of care and support for caregivers.

.Compliance with drugsMedication adherence is a huge challenge in health care, especially for people with complex drug regimens. [56] Voice-based solutions that remind the patient of the dose and schedule, and offer improved content relevant membership and contextually reduce the risk of abuse.

.Mental HealthPowerful voice applications for patients with mental health issues such as depression can provide self-help support and a way to assess cognitive function over time. [57]

.Collection of dataThe more general collection of data such as quality of life indicators, functional assessment variables, level of pain, and other health indicators can be simplified through voice assistant technologies. [58] [59]

III.2.2 Snips- personalize your voice assistant with confidentiality

Snips, an artificial intelligence voice platform for connected devices, people can now download Snips SDK, its open source, and connect with a Raspberry Pi, a speaker and a microphone.

It is a startup based in France involved in voice assistant technology.

Snips does not use Alexa Voice Service or Google Assistant SDK, the company builds its own voice assistant so that it can be integrated on the devices. And the best part is that it does not send anything to the cloud because it works offline. Because the personal standby word detector is locally formed, queries are stored on the device rather than on the cloud. This model confirms our commitment to a voice assistant “Privacy Premier”, so that personal data will not be shared with advertisers, governments or anyone else. So, this voice assistant is “private by design” and can work without an Internet connection. [60]

If we want to understand how a voice assistant works, we can divide it into several parts. First, it starts with a password. Snips has a handful of default passwords, like “Hey Snips”, but you can also pay the company to create their own password.

For example, if you are building a multimedia robot called Keecker, you can create a special word “Hey Keecker”. Snips then uses deep learning to accurately detect when someone is trying to talk to your voice assistant.

The second part is automatic speech recognition. A voice assistant transcribes your voice into a text request. Popular family assistants usually send a small audio file with your voice and use servers to transcribe your request.

Shears can transcribe your voice into text on the device itself. This works on anything that is more powerful than a Raspberry Pi. For now, Snips is limited to English and French. You can use a third-party automatic voice recognition API for other languages.

Then, Snips must understand the user’s request. The company has developed natural language skills. But there are hundreds, if not thousands, of different ways to ask a simple question about the weather, for example.

That’s why Snips today launches a data generation service. The interface looks like Automator on macOS or Workflow on iOS. We can define some variables, such as “date” and “location”, we can define if they are mandatory for the query and you enter some examples.

But instead of manually entering hundreds of variations of the same query, you can pay from $ 100 to $ 800 to allow Snips to do the work for you. Startup manually checks your request and publishes it on Amazon Mechanical Turk and other crowdsourcing markets. Finally, Snips cleans your dataset and sends it back to you.

The user can either download and reuse it in another chatbot or voice assistant, or use it with the Snips voice wizard. One can also make one’s own capacity public. Other Snips users can add this feature to their own wizard by browsing a repository of pre-trained capabilities.

Snips also injects a business-to-customer element into its business, so it will compete directly with the speakers of Amazon Echo and Google Home. And the Snips start-up will deploy the Snips AIR Base products and Snips AIR Satellites .

The base will be a good old smart speaker, while satellites will be tiny portable speakers that users can put in all their rooms. During this time, there will be a market where developers will be able to generate skills and to publish them. The market will run on a new blockchain known as the AIR blockchain.

The market will take AIR to buy more skills. With such AIR, users will also be able to produce training data for voice commands.

III.2.3 Mycroft – does not collect or monetize your data

Mycroft is an open source voice assistant that can be expanded. MyCroft AI integrates natural language processing, speech-to-text and text-to-speech speech synthesis to create a powerful conversational experience. MyCroft AI is an open source software licensed under the GNU General Public License version 3.0, which means it can be freely remixed, extended and enhanced to meet unique needs. The MyCroft AI platform focuses on voice recognition, allowing any device to turn it into a smart personal assistant, which can perform a variety of tasks. MyCroft AI takes advantage of new technologies, which allow users to collaborate, share ideas and help each other to optimize the use of applications. MyCroft AI can be used in any project, whether it is a scientific project or an enterprise software application. [61]

MyCroft’s cross-platform model allows seamless integration wherever and whenever you want. The app learns with you and improves your usage experience over time. It also processes and responds to all kinds of queries. It’s a handy wizard that gives you information about your daily commitments such as time and date, weather, and helps you perform your daily tasks such as setting up alarms or playing your favorite playlist. MyCroft AI Mimic is a fast, lightweight voice synthesis engine that takes text and reads it aloud to create a high-quality voice. MyCroft AI provides a software library for converting natural language into data structures. MyCroft AI is also developing OpenSTT; a feature that converts speech into text for high-accuracy, low-latency speech conversion.

MyCroft AI uses Linux, an open-source operating system and can run on any device, be it on a desktop, tablet or mobile phone. The documentation to guide you in the implementation is available on the website of the application.

Unlike more familiar models such as Amazon Echo and Google Home, Mycroft’s goal is to build an open source alternative to the IAI of the big tech giants. assistants, and one who promises to protect your privacy in the process. [62] Companies like Google and Amazon want to confine you to an ecosystem they control and then monetize you.

The big differentiator, though, is privacy. Although Mycroft can answer questions like all smart assistants, this data will not be sold to advertisers and will be returned as ads. In an ideal world, this means that the information you obtain is objectively the best information for you; Mycroft will not store any user voice data on its servers unless the user explicitly agrees to allow the user to do so in order to improve its content. voice recognition efforts. If you agree, you can change your mind at any time. The goal of Mycroft is to be more transparent than companies like Google and Amazon that, at least from now on, make it difficult to massively delete your voice files – and warn that this can affect the usability of your devices.

For more clarity, Mycroft will treat the listener in the same way as its biggest rivals. The speaker will listen to a wake-up message using a processor on the device and will not start downloading audio until it has been triggered.

The distinction is what happens to the audio after it goes to the cloud. Although Amazon and Google allow users to review and delete their voice recordings, neither company offers a way to automatically delete data over time. And when users attempt to erase mass data, an option available only on the Google and Amazon websites, and not on their mobile voice assistant applications, the two companies do not recommend it. A pop-up on Amazon’s website indicates that deleting “may degrade your experience,” while Google displays a pop-up that says it can “make Google services more useful” if you keep these examples of voice in your files. [63]

But the downside to MyCroft’s uncompromising privacy stance is that it limits the company’s ability to improve speech recognition. Instead of training his AI from scratch, Mycroft uses open data sets from voluntary voice recordings and other open sources. This is equivalent to a fraction of what Google and Amazon collect. [64] There are significant disadvantages to creating a voice assistant without massive amount of personal voice data.

Generally, for services like voice assistant, the more data you have, the easier it is for you to develop a truly effective and truly responsive model.Open Source means that Mycroft is not confined to an ecosystem, so it is compatible with any e-mail service, social feeds, music streaming you want. If you are a person with an iPhone, iPad, Mac and Apple TV, it’s best to stay with HomePod – but if you want a little more flexibility, this might be the solution for you.

Plus, not only can Mycroft talk, but it can also display graphics and widgets so that the user can better interpret what the device is trying to say.

III.2.4 Orbita Voice-Building Multimodal Voice Experiments for Health

Orbita Voice is the first platform to integrate intelligent voice assistant technologies, such as Amazon Alexa and Google Assistant, discussion robots and other conversational interfaces into healthcare applications. [65]

Orbita’s Health Cloud platform brings data from connected health care and care appliances into collaborative care experiences that greatly enhance patient engagement and care management. [66]

Orbita solutions do not need to have their own speech recognition technologies or other basic AI technologies, but focus their core competencies on business logic clarification and knowledge mapping for services medical. [67] In addition, it is the ability to integrate intelligent voice, instant messaging tools, natural language understanding and other related technologies.

Take a speech example with Orbita Voice below:

We can see that human-machine interaction in Orbita can have at least the following new capabilities:

-The instant request. Combining time, location, personal conditions and other factors, the chat robot provides a more accurate and efficient instant query function. In this example, robot will automatically confirm the time, users will not need to type them, robot will immediately note the type of medicine, but will also inform the amount. In addition, the patient does not remember what type of pill was present, and robot described the shape of the tablet.

-Active demand. Unlike previous interactions between humans and machines, discussion robots will also follow users in a timely manner. In this example, robot is actively asking the user whether to send an image to compare. This is also difficult to implement in the human-machine interaction interface of the past.

-The recording automatically. In the past, many health-related applications were needed to record details such as diet, fitness, and medication. The previous text entry method was very tedious and it was difficult to adhere to it on a day-to-day basis. after day. The Orbita mode no longer needs to be manually entered, all you need is to be explained to the robot, and processes the recordings in the background. In this example, the user tells Bot that he has taken good medication, after hearing this, rotbot automatically saves the database data.

Hardware devices that can support intelligent voice include not only intelligent audio, but also intelligent hardware such as smart phones, portable devices, and home medical devices. In addition, the voice mode is not a mandatory mode, but an optional mode, apart from speech, Orbita can still use interactive discussion robots.

Orbita is also an Omni-Channel visual designer. Write once and deploy it on Amazon Alexa, Google Assistant and other voice and chat services. Speed ​​development can be speeded up with an intuitive graphic designer. manage multiple voice applications in one place.

III.3. Communication networks

Developments in low power communication technologies, miniaturized sensors and actuators as well as the growing penetration of the internet, tablets and smartphones are driving us into the new era of IoT. [68] The Internet of Things can enable new ways of living by connecting the physical world to the digital computing platform through intelligent sensing and actuation devices and appropriate communications technologies.

Therefore, the concept of IoT can be exploited in a wide range of applications for e health and home care.

Traditional dwellings are not usually designed to monitor the physiological conditions and occupant activities by itself. In contrast, a smart home is a traditional home with smart devices and modern communication technologies that can facilitate automatic monitoring of the home environment, safety and health status of the occupants. [69]

However, in order to achieve broad acceptance among users, smart homes must be affordable. Therefore, low-power and efficient communication technologies and public networks, as well as low-cost devices are essential for smart homes. In addition, several key technological challenges such as the total interoperability between interconnected devices, the high degree of accuracy and precision, the limitation of processing resources and the security of confidentiality and information must be addressed. [70]So we must first build a state of the art on different communication techniques and make the comparison.

Wired communication networks

Home networks of wireless communication

III.4. The innovation of IOT-Matrix Creator

III.4.1 Introduction Matrix Creator

The Internet of Things (IoT) is everywhere. Each connected device is marketed as “smart.” The problem is that most of these IoT devices are basic and simple: integrating the infrastructure of “connected objects” as sensors by making connectivity in all the devices in sight.

Now, devices and devices embedded in this ecosystem must start talking to each other. Using a combination of snap-in hardware and open-source software, MATRIX aims to create a real Internet of Things, as opposed to a lot of things that connect.

MATRIX Creator is a complete development board with sensors, wireless communications and an FPGA. MATRIX Creator is designed to provide every manufacturer, hobbyist, and developer around the world with a comprehensive, affordable, and easy-to-use tool for creating simple to complex Internet of Things (IoT) applications. [101]

The MATRIX device and all new technologies provide a converged hardware / software platform that gives developers the freedom to experiment with IoT applications and create an open-source ecosystem around MATRIX.

The hardware itself is a quad-core 64-bit 1.2GHz ARM Cortex A53 with 32GB of memory, a 5-megapixel 1080p camera, WiFi 2.5 and 5 GHz, eight microphones. MATRIX CREATOR can communicate with ZIGBEE, Z-WAVE, NFC, BLUETOOTH networks, it contains the sensor of temperature, pressure, UV, movement and orientation, it is also compatible with Raspberry Pi, Debian and Android Things. MATRIX also sports a rainbow colored LED ring around the circular device called “Everloop”.

On the software side, the open-source gesture.ai SDK allows developers to code gesture triggers in any camera-supported application. MATRIX also comes with three IoT applications the company has grown:.MATRIX Automation for connection and control of household or professional appliances;.MATRIX Security using computer vision to monitor remote locations;.MATRIX MIA (MATRIX Intelligent Assistant) uses voice and gesture commands to control organization and productivity at Alexa and google assistant.

MATRIX democratizes and opens the development of IoT applications thanks to a centralized experience. This “Intelligent Application Ecosystem” is an IoT hub where developers can create and download IoT applications for individuals and businesses. Theoretically, you can create almost anything with MATRIX Creator.

III.4.2 Layers of Matrix Creator programming

The MATRIX platform adds powerful features to Raspberry Pi, depending on the background and type of application, you can choose the layer that best fits the needs:

MATRIX OSMatrix OS (MATRIX Open System) is the highest level of abstraction integrating MATRIX hardware via MATRIX CORE.It is an open source software for hosting IoT applications. MATRIX Open System runs on node.js and, initially, applications will be written in JavaScript. The current workflow of this platform to deploy and install applications in your MATRIX device, from a separate computer, anywhere in the world.

This layer requires an online connection and provides an easy-to-use integrated IoT environment, including remote application deployment, mobile interface, online dashboard and App Store.

MATRIX COREMATRIX CORE is an abstraction layer for MATRIX HAL and the foundation for MATRIX OS. This layer uses Protocol Buffers & ZeroMQ to communicate with MATRIX device which makes device information accessible via high-level interfaces. Applications for the MATRIX device can be programmed with any language that supports these tools. Supports more than 40 different languages ​​thanks to Protocol Buffers: C ++, Python, Ruby, PHP, Java, etc.

MATRIX HALHAL (Hardware Abstraction Layer) is the lowest level abstraction layer for MATRIX Creator drivers. You can interface directly with HAL or use top-level components such as Matrix CORE and MATRIX OS.Interacts with kernel modules using C ++ drivers, allowing it to access the sensors and components available on your device.

III.5. Central control platforms

IntroductionWith an ever increasing number of devices available to help achieve e-health at home and automate the home. Whether you want to monitor your health or control your lighting system remotely, protect your home against theft, fire or other threats, reduce your energy consumption, there are countless devices to your home. disposition.

But at the same time, many users are worried about the security and privacy implications of introducing new devices to their homes. They want to control who has access to the vital systems that control their devices and record every moment of their daily lives. Even if you allow a device to communicate externally, it is only accessible to explicitly authorized users.

Being able to fully understand programs that control the home means that you can see and if necessary modify the source code that runs on the devices themselves.

While connected devices often contain proprietary components, we must ensure that all devices are connected to each other and interface with the hub. So we need an open source platform.

There are many choices, with options for everything, I focused mainly on three Home Assistant, OpenHAB and Domoticz.

Home AssistantHome Assistant is an open source home automation platform designed to be easily deployed on almost any machine that can run Python 3, from a Raspberry Pi to a network-attached storage device, and even with a Docker container for on-premise deployment. other systems a breeze. [102] It integrates with a large number of open source offers as well as commercial offers, allowing you to link as IFTTT [103], Amazon Echo, Google assistant, snips, Z-wave to control hardware locks to lights .

This platform offers easy-to-use user interfaces for all mobile devices. A great feature of Home Assistant is that it does not store all your private information on the cloud, so you can be sure everything is locked and secure. They also have a lively community, so you can find the help you need – anytime!Home Assistant is released under an MIT license [104], and its source can be downloaded from GitHub.

OpenHABOpenHAB (Open Home Automation Bus) is one of the most popular home automation tools among open source enthusiasts, with a large user community and a large number of supported devices and integrations. [105] Written in Java, openHAB is portable on most major operating systems and works well on the Raspberry Pi. Supporting hundreds of devices, openHAB is designed to be device independent while making it easy to add their own devices. own devices or plugins to the system. OpenHAB also offers iOS and Android apps for device control, as well as design tools to help you build your own user interface for your home system.

OpenHAB offers a powerful rules engine and transparent user interfaces for a variety of platforms, and is one of the best known and most popular of the open source community because of its ease of use. smart home also has a large community of loyal supporters with whom users can connect to solve problems or share ideas, which is precisely the starting point of open source software!OpenHAB is ideal for a user who has some experience with the software but may not know much about home automation. You can find the openHAB source code on GitHub under the Eclipse Public License.

DomoticzDomoticz is a home automation system with a fairly large library of supported devices, ranging from weather stations to smoke detectors to remotes, and a large number of additional third-party integrations are documented on the project website. It is designed with an HTML5 interface, making it accessible from desktop browsers and most modern smartphones, and is lightweight, running on many low-power devices like the Raspberry Pi. [106]

This open source system is also designed to integrate different devices from different vendors. It offers the same notifications as any proprietary smart home software, and it supports a huge amount of home automation devices ranging from switches to sensors. all types. It also has an automatic learning feature that makes installation easier.The website features an active update and plug-in section with a ton of third-party add-ons that further increase functionality. This open source software is also designed to work in light and use very little energy.Domoticz is perfect for a user who may not be very familiar with software coding, but who understands how a home automation system works and wants to modify their system.Domoticz is written mainly in C / C ++ under GPLv3, and its source code can be browsed on GitHub.

Personally, Home Assistant is the most technically scalable platform with all the software developed in Python 3, operating asynchronously and with very low requirements in terms of specifications. We can order our home lights and window coverings using a PLC connected to Home Assistant via the MQTT protocol and receive information from multiple sensors using a Z-WAVE network. It also connects to IP cameras to retrieve motion detection events and can notify us via various services. The possibilities are enormous.

IV.The design and a proposal of mockup

IV.1 Philosophy of conception

There are many technologies and equipment for smart home support and e-health. Which one should we take? But at the beginning, in principle, we think that a home e-health system can include three layer: Communication Network, Decision Making Platform and Services.

Communication networkAll sensors and other devices are connected to the home with central communication via a communication network, which forms the first layer of the architecture. All data such as physiological signals, environmental signals measured by the sensors and control signals are transmitted to the central computing node on a communication medium.

We can use multiple sensors to collect data about the home environment, such as lighting level, temperature, pressure, gas leaks and occupant activity or location. Physiological parameters such as blood pressure, body weight can be measured using portable sensors.

Portable medical sensors can be connected in a body sensor network, where the central node is connected to all environmental sensors via the wireless sensor network. All smart home sensors and equipment are connected to form a local area network or personal networks and to provide data communication within the smart home. The central decision-making platform can communicate with all sensors and devices using the network to collect data or send feedback to perform necessary actions as needed.

Decision-making platformThe second layer of architecture is responsible for computing and decision-making, thus functioning as the brain of the system. This layer will be equipped with a computer system such as a custom-built processing node based on microprocessors. It collects sensor and actuator data over the wireless sensor network, processes and analyzes the measured data, and sends feedback back to the user (s). It can also store measured data, display results for the user, and perform prediction algorithms. Predictive algorithms can exploit the characteristics of artificial intelligence and use deep learning and machine learning techniques to learn and develop models for the home environment as well as behavioral and physiological models of occupants. We can equip stand-alone devices with smart sensors to monitor and evaluate the activity and overall health of residents.

The computing platform can make predictive decisions about the home environment or the health status of the occupant based on information received from multiple sensors. The adoption of Artificial Intelligence will also enable this platform to leverage robotics to control smart home devices and provide occupant services automatically with continuous improvements in accuracy over time.

This layer can transmit the measured data, the physiological or environmental parameters on the Internet, thus functioning as an access gateway to the remote installation. This platform continuously monitors and evaluates the measured physiological or environmental data. If an anomaly in the home environment or in the vital signs of the user’s health condition is detected, it can trigger an alarm or send alert messages to the service providers in the form of a voice call, SMS or e-mail.

ServicesThe top layer of the architecture includes the services provided to the user by the service providers. These services may be associated with occupant health, the environment, the safety or security of the home and residents. These services could be local services or remote services.

Simple services such as wireless lighting controls, voice assistance, fall detection, etc. More complex such as telemedicine, social networks for chronic disease management. The services provided may be adapted to the requirements of the occupants based on the level of medical care or the required safety and security.

In a smart home, the gateway platform functions as the main service provider, for example, by activating the actuators needed to control the home environment, lighting, in the case of automatic delivery of drag.

The gateway system can control smart devices to provide occupants with better services. The bridge can learn and keep a continuous record of the physiological conditions of the occupants. It can also monitor the home environment and detect any dangerous situation such as the presence of smoke or gas leakage using environmental sensors installed at different locations in the home.

The secondary service provider is the central hub of all subscribed smart homes and is responsible for the management, maintenance, connectivity and information security of the network and smart home systems. It constantly monitors alarms or emergencies and immediately informs other third party services such as the emergency medical service, carers, etc.

IV.2 Realization

We ask the requirements and constraint of the system designIt requires the network is simple and can be adapted for use in small places such as home and retirement homes.The network communication protocol requires low power consumption and high security.You can collect physiological parameters that reflect the health status of the elderly (EGC, heart rate, temperature, etc …)The system must be able to have more functionality by combining voice, video, data equipment.Data collection nodes must have a low impact on the normal lives of older people.It also requires that the terminals have a user-friendly man-machine interface, a simple operation and a data storage. And show the features, and have a good scalability.

Older people in the home can control different smart appliances (lamps, switches, power outlets) and measure blood pressure, heart rate, ECG etc. We can provide the services of health monitoring, Emergency Rescue Emergency, Alarm, Ambulance Remote, Health Advice.

The journey of a thousand miles begins with a step. So we want to first design and make a small test model with figure below.

I used Z-WAVE as the home network and the communication network layer of this test because it is quite affordable and flexible enough to be installed in virtually any personal technology, such as blood pressure monitors and scales. Therefore, Z-Wave is sufficiently powerful and reliable for critical healthcare applications. Physicians and health care institutions are increasingly relying on technologies that can help them care for more patients and provide them with more reliable and faster access to vital information while on the move. In a short time, Z-Wave will support these technologies, and a large part of our health systems.

While technology plays a central role in caring for the elderly and disabled, Z-Wave is an important bridge between doctors, caregivers and patients, namely communication. There is also more and more ZWAVE products on the market.

Considering powerful features of MATRIX CREATOR, I would have liked to make the Z-WAVE controller and voice assistant in a MATRIX device meme. Matrix creator and Home assistant can be taken as the second layer of decision-making platform.

For the service layer, we want to prevent detect the environment of the house, detect the fall, detect the presence of people, control different smart equipment with voice, smartphone and computer.

One can also measure blood pressure, pulse rate and automatic saving of results. All measurement results could be uploaded to the server for analysis and monitoring by Z-Wave access.

V.Implement the “mockup”

VI.Product order

After fully studying the documentation, I did a research on the different manufacturers of Z-WAVE products via the internet to see what products were providing because we had to order some products to test the performance and realize the installation. I realized that there was a wider choice of products among 700 Z-WAVE alliance members and 2400 product references. I then concentrated on studying the products of these three companies FIBARO, ZIPATO and Xiaomi by considering factors appearance, performance and value for money products.

After studying the different products and considering the price factors, quality but also the expansion of the system, I finally decided to take the MATRIX CREATOR system to achieve controller Z-WAVE but also as gateway IP / Z-WAVE. We also wanted to use MATRIX CREATOR to realize the vocal assistant and the vision management. So we ordered the quantity.

Considering that the products to measure the physiological parameters of health, a product JTB-1001-02 Z-Wave Blood Pressure Monitor is very interesting.This product includes the measurement of blood pressure, pulse measurement and automatically save the result. It also has arrhythmia alarm and audio notification. He installed multi-level sensors and he had multichannel associated with diastolic pressure / contraction pressure / pulse heart rate. The design provides clients with many years of reliable service. The reading taken by the monitor is equivalent to those obtained by a qualified observer using the cuff and stethoscope auscultation method. All measurement results can be downloaded to the health care server for analysis and tracking via Z-Wave access. But by the time I made the order and contacted this manufacturer, there is no inventory on this product.

Fibaro motion sensor is a stylish, well made and reliable ZWave sensor.This multi-sensor is pretty decent.It can detect motion, temperature variation, light and vibration. This is because of the accelerometer installed inside its small compact body. The Fibaro comes with an alert system that lovers of smartphones or tablets will surely love. If you have been away, you can easily monitor your home by synchronizing the device on a phone or tablet. You can also receive notifications by e-mail.You can trigger the alarm or activate specific preset scenes remotely when motion detection or light detection is triggered.If there are pets at home that could trigger false alarms, we can easily adjust the level of sensitivity to distinguish pets from people.If your problem is falsified, you do not have to worry because the device is equipped with an anti-tampering mechanism. sabotage that triggers alerts when someone moves the device.

And we also thought that put the traditional home appliances in the network.The product Fibaro FGWPE-102 ZW5 EU Socket and EVERSPRING Module socket can play this role. The FIBARO wall socket is an ultimate and intelligent remote plug adapter. This wall socket can be applied wherever you want to control electrical devices in the Z WAVE network, while monitoring energy consumption in a convenient and maintenance-free way. The results can be seen in an easy-to-read pie chart and graphs not only of the amount of energy that the appliances are taking, but also of the plug itself. The EVERSPRING module socket AN145 allows remote control an E27 lamp via Z-Wave controls.It can be controlled by a remote control, PC software, or any Z-Wave controller on your network.Each Z-Wave module works as a wireless repeater with the other modules , to ensure total coverage of the home.

The detailed list and price of the products purchased can be seen in the table below:

V.II Realization

Matrix Creator TestAfter receiving the products ordered, the first thing to do is to test MATRIX Creator. As mentioned earlier, MATRIX Creator has three levels of the layer, so I tested these three levels: MATRIX OS, MATRIX CORE and MATRIX HAL.

MOS contains a Command Line Interface (CLI) tool to control and manage your MATRIX devices. To install the tool, we must execute the command in the terminal of the personal computer. Once installed, the CLI tool must be configured by registering and logging in to a MATRIX Labs account.

Once the device is created, a set of unique credentials will be generated. Register these credentials as they are needed to link the MATRIX Labs account to the MATRIX device.

Access the Raspberry Pi terminal via an SSH session. Here I used the software called Mobaxterm to get Raspberry control. And I used Advanced IP Scanner to find the Raspberry IP address.Next, connect the Device To MATRIX Labs account, install and run the application. Here, I installed the HelloWorld app from the MATRIX App Store and tested it.

The output of the result as the image below.

To test the MATRIX CORE layer, the MATRIX CORE package and ZeroMQ must be installed. And we have used a Python project (Appendix I.1) to test it. We have the result as the picture below.

To test the MATRIX HALL layer, you can download the package directly to the site. Basic examples can be found in the demo directory inside the repository we did the installation. We have the result as the picture below.

Realization Z-WAVE controller and networksTo make a Z-WAVE controller, use the MATRIX CORE layer and the MATRIXIO kernel module. The Zwave abstraction layer for MATRIX Creator can be used via 0MQ. Protocol buffers are used for data exchange.

We can also use MATRIX CORE to query the MATRIX Creator sensors and control the MATRIX Creator from any language that supports protocol and 0MQ buffers. Connections to MATRIX CORE can be made from localhost (127.0.0.1) and from remote computers on the same network.

It is necessary to first program and compile the MATRIXIO kernel module (Annex I.2), it is also necessary to build a Z / IP gateway to facilitate the control of Z-Wave devices from the IP network (Annex I.3), then do the initial configuration of Zwave and Matrix CORE (Annex I.4).

After these schedules, we need to know that there are some steps to work on: start the Z-WAVE / IP gateway, find Z-Wave services in the network and start include the node. But at first I was blocked by the Matrix card and Z-Wave did not work even though all the programming was done. That made me over a month to debug the problems and I found the cause.

The Z-Wave is connected to the FPGA using a serial port. But the serial port of MATRIX is used by default zigbee. The Z-WAVE / IP configuration is assigned to the ttyACM0 gate. We can not use ZIGBEE and Z-WAVE at the same time in Matrix Creator. So we need the port to be z-wave. You have to type the commands (Annex I.5) to change the Z-WAVE gate at the ttyS0. Until then, we can add Z-wave devices and control them under Mobaxterm by typing the commands. (Annex I.6)

Connection with HOME ASSISTANT and configurationThe control of ZWAVE in mobaxterm with SSH is not conducive to data analysis and does not favor non-professionals using the network. So we used HOME ASSISTANT.

After the installation of HOME ASSISTANT, it is important to edit the configuration.yaml file (Annex I.7) in the home assistant directory. We can divide into several groups, according to the needs, the living room, the rooms, the offices, cooking, and control of different equipment and instruments.

With debugging, all Z-WAVE can be added to HOME ASSISTANT, which can control lights, switches and network instruments via computers and mobile phones.

But it shows no data for sensors including FABIRO MOTION SENSOR on the HOME ASSISTANT platform. There is no content to configure for “node group associations” and “node configuration parameter”.

Regardless of the device I choose and they show the “Group”, “Configuration Setting”, there is no data displayed. And the sensor was closed even though I clicked button B three times. (Annex II.1) I spent more than two weeks finding a solution for this show, but no answer. So I turn to the other direction.

Realization VOCAL ASSISTANT with Matrix CreatorAt first, I would like to do google assistant in the same Matrix creator Z-WAVE.But it shows the errors. (Annex II.2) So, I used another new MATRIX Creator.In order to allow the Google Assistant software to To access the microphones of MATRIX Creator, we must install the Matrix kernel module and install the Google SDK Assistant. (Annex I.8) After the test, Google Assistant function well and can answer our questions.

Under the same Matrix creator, I installed and tested the other SNIPS voice assistant. (Annex 1.9) It works well too. Compared to google assistant, it has fewer functions, and it takes longer to manually enter questions and answers.The voice recognition capability is also lower than that of GOOGLE ASSISTANT. After that, I spent three weeks trying to control the lights with sound, through the SNIPS or GOOGLE ASSISTANT, but I did not succeed.

My model realization has temporarily arrived here.

VI. Conclusion

This reasearch can give you the opportunity to discover the service of home care and e-health.

You can have a preliminary understanding of the internet of things and the MATRIX creator application in the field of e-health and home support.

We can create our own Assistant Artificial Intelligence Assistant Google Assistant, Alexa, Snips with Matrix Creator and convert to each other on the same device. In this implementation on the website, only the conversion between google assistant and snips is implemented.

You can also use different types of communication networks: ZIGBEE and ZWAVE. We can also use MATRIX CREATOR to build different intelligent communication networks in order to facilitate the realization of an eco e-health system at home. For the beginners to the MATRIX CREATOR system, it is of course complicated. We realized the Z-WAVE network and did the troubleshooting by spending almost two months.

We can say that MATRIX Creator is a powerful tool and platform ready for IoT that allows to build computer vision, home automation, robotics and other e-health solutions, which is integrated with 15 sensors .

MATRIX Creator can satisfy both novices and experts, offering a simplified Linux operating system that runs on the PI and is programmable in JavaScript, Python and C ++.Of course, we can go further and better, we can build more advanced detection applications with face, gesture and object detection, can access a pre-existing computer vision library for free.

This reasearch allow you to know the various functions that are commonly found in the e-health field. It also allowed you to have the initial capacity to conceive and propose innovative solutions in the home-based smart health system.

Annex

AnnexI.1 A simple app.py project to test MATRIX Creator

## Set Initial Variables ##

import os # Miscellaneous operating system interface

import zmq # Asynchronous messaging framework

import time # Time access and conversions

from random import randint # Random numbers

import sys # System-specific parameters and functions

from matrix_io.proto.malos.v1 import driver_pb2 # MATRIX Protocol Buffer driver library

from matrix_io.proto.malos.v1 import io_pb2 # MATRIX Protocol Buffer sensor library

from multiprocessing import Process, Manager, Value # Allow for multiple processes at once

from zmq.eventloop import ioloop, zmqstream# Asynchronous events through ZMQ

matrix_ip = ‘127.0.0.1’ # Local device ip

everloop_port = 20021 # Driver Base port

led_count = 0 # Amount of LEDs on MATRIX device

# Handy function for connecting to the Error port

from utils import register_error_callback

## BASE PORT ##

def config_socket(ledCount):

# Define zmq socket

context = zmq.Context()

# Create a Pusher socket

socket = context.socket(zmq.PUSH)

# Connect Pusher to configuration socket

socket.connect(‘tcp://{0}:{1}’.format(matrix_ip, everloop_port))

# Loop forever

while True:

# Create a new driver config

driver_config_proto = driver_pb2.DriverConfig()

# Create an empty Everloop image

image = []

# For each device LED

for led in range(ledCount):

# Set individual LED value

ledValue = io_pb2.LedValue()

ledValue.blue = randint(0, 50)

ledValue.red = randint(0, 200)

ledValue.green = randint(0, 255)

ledValue.white = 0

image.append(ledValue)

# Store the Everloop image in driver configuration

driver_config_proto.image.led.extend(image)

# Send driver configuration through ZMQ socket

socket.send(driver_config_proto.SerializeToString())

#Wait before restarting loop

time.sleep(0.05)

## KEEP ALIVE ##

def ping_socket():

# Define zmq socket

context = zmq.Context()

# Create a Pusher socket

ping_socket = context.socket(zmq.PUSH)

# Connect to the socket

ping_socket.connect(‘tcp://{0}:{1}’.format(matrix_ip, everloop_port+1))

# Ping with empty string to let the drive know we’re still listening

ping_socket.send_string(”)

## ERROR PORT ##

def everloop_error_callback(error):

# Log error

print(‘{0}’.format(error))

## DATA UPDATE PORT ##

def update_socket():

# Define zmq socket

context = zmq.Context()

# Create a Subscriber socket

socket = context.socket(zmq.SUB)

# Connect to the Data Update port

socket.connect(‘tcp://{0}:{1}’.format(matrix_ip, everloop_port+3))

# Connect Subscriber to Error port

socket.setsockopt(zmq.SUBSCRIBE, b”)

# Create the stream to listen to data from port

stream = zmqstream.ZMQStream(socket)

# Function to update LED count and close connection to the Data Update Port

def updateLedCount(data):

# Extract data and pass into led_count global variable

global led_count

led_count = io_pb2.LedValue().FromString(data[0]).green

# Log LEDs

print(‘{0} LEDs counted’.format(led_count))

# If LED count obtained

if led_count > 0:

# Close Data Update Port connection

ioloop.IOLoop.instance().stop()

print(‘LED count obtained. Disconnecting from data publisher {0}’.format(everloop_port+3))

# Call updateLedCount() once data is received

stream.on_recv(updateLedCount)

# Log and begin event loop for ZMQ connection to Data Update Port

print(‘Connected to data publisher with port {0}’.format(everloop_port+3))

ioloop.IOLoop.instance().start()

## START PROCESSES ##

if __name__ == ‘__main__’:

# Initiate asynchronous events

ioloop.install()

# Start Error Port connection

Process(target=register_error_callback, args=(everloop_error_callback, matrix_ip, everloop_port)).start()

# Ping the Keep-alive Port once

ping_socket()

# Start Data Update Port connection & close after response

update_socket()

# Send Base Port configuration

try:

config_socket(led_count)

# Avoid logging Everloop errors on user quiting

except KeyboardInterrupt:

print(‘ quit’)

Annex I.2 ​​Compile the MATRIXIO Kernel Module

MATRIXIO Kernel is the kernel driver for MATRIX Creator. This driver only works with the current raspbian kernel. To return to the current use of the Raspbian kernel:sudo apt-get install –reinstall raspberrypi-bootloader raspberrypi-kernelInstall dependencies# Add a repo and a keycurl https://apt.matrix.one/doc/apt-key.gpg | sudo apt-key add –echo “deb https://apt.matrix.one/raspbian $ (lsb_release -sc) main” | sudo tee /etc/apt/sources.list.d/matrixlabs.list# Update packages and installsudo apt-get updatesudo apt-get upgrade# Installation MATRIX Pacakagessudo apt install matrixio-creator-init# Kernel installation packagessudo apt-get -y install raspberrypi-kernel-headers raspberrypi-kernel git# To restartsudo reboot# Compilinggit clone https://github.com/matrix-io/matrixio-kernel-modulescd matrixio-kernel-modules / srcmake && make installAfter that, you must overlay the setup programAdd /boot/config.txtdtoverlay = matrixioThen restart the system. It allows to activate the module MATRIXIO Kernel.

Annex I.3 build a Z-WAVE / IP gateway

Run the following commands:$ sudo apt-get update$ sudo apt-get install cmake libavahi-client-dev libxml2-dev libbsd -dev libncurses5-dev libncurses5-dev gitUsing Raspbian Jessie install:sudo apt-get install –yes libssl-devUsing Raspbian Stretch install:sudo apt-get install –yes libssl1.0-devThen:$ git clone https://github.com/Z-WavePublic/libzwaveip.git$ cd libzwaveip$ mkdir build$ cd build$ cmake ..$ make

Annex I.4 The Zwave and Matrix CORE configuration

Initial configuration of Zwave# Add a repo and a keycurl https://apt.matrix.one/doc/apt-key.gpg | sudo apt-key add –echo “deb https://apt.matrix.one/raspbian $ (lsb_release -sc) main” | sudo tee /etc/apt/sources.list.d/matrixlabs.list# Update packages and installsudo apt-get updatesudo apt-get upgrade# Install MATRIX Creator Initsudo apt-get install matrixio-creator-init matrixio-kernel-modules# Install Zwave utilitiessudo apt-get install matrixio-zwave-utils

Run zwave_conf to configure the ZM5202 in MATRIX Creator$ zwave_confThen restart the system. This process must be run once. The ZM5202 will retain this configuration.installsudo apt-get install matrixio-malos-zwavesudo rebootRunning as a serviceAt this point, at the next start, matrixio-malos-zwave will work as a service called: status matrixio-malos-zwave.service.sudo systemctl status matrixio-malos-zwaveUpgradesudo apt-get update sudo apt-get upgradesudo rebootManual start# to run manually, just type `malos`malos_zwaveWe can check if the zipgateway works with $ more /tmp/zipgateway.log

Annex I.5 Changing the door from Z-WAVE to ttyS0

git clone https://github.com/matrix-io/matrix-malos-zwave.gitcd matrix-malos-zwaveblob cdsudo ./zwave_setup.bash copy

Annex I.6 Using ZWAVE with Matrix Creator

To run the Z / IP Gateway~ / libzwaveip / build / reference_client -s fd00: yyyy :: 3Find Z-Wave clients in the network -can find it using the list operation

 Just include a node (device) with the addnode operation

 An example of order to turn on FIBARO wall plug(ZIP) send “Binary Switch [e100778e0a00]” COMMAND_CLASS_SWITCH_BINARY SWITCH_BINARY_SET ff

Annex I.7 HOME ASSISTANT Configuration

configuration.yamlhomeassistant: # Name of the location where Home Assistant is running name: Home # Location required to calculate the time the sun rises and sets latitude: 48.6748 longitude: 2.3006 # Impacts weather / sunrise data (altitude above sea level in meters) elevation: 0 # metric for Metric, imperial for Imperial unit_system: metric # Pick yours from here: http://en.wikipedia.org/wiki/List_of_tz_database_time_zones time_zone: Europe / Pariscustomize: sensor.alarm_type: friendly_name: SENSOR ACTIONS hidden: false # Customization file customize:! include customize.yaml# Show links to resources in frontendintroduction:# Enables the frontendfrontend:# Enables configuration UIconfig:http:# Discover some devices automaticallydiscovery:# Allows you to go to the frontend in enabled browsersconversation:# Enables support for tracking state changes over timehistory:# View all events in a logbooklogbook:# Enables a map showing the location of tracked devicesmap:# Track the sunsun:# Weather predictionsensor: – platform: yr# Text to speechtTS: – platform: google# Cloudcloud:# Turn on Z-WAVE from HOME ASSISTANT to the ttyS0 doorZwave: usb_path: / dev / ttyS0group: default_view: # default homepage is HOME page view: yes # Indicates if pagination is displayed at the beginning of the page entities: # Peripherals and groups under the first tab – sun.sun – sensor.yr_symbolgroup:! include groups.yamlautomation:! include automations.yamlscript:! include scripts.yamlConfiguration groups.yamlThe apartment: # This is the setting of the whole ROOMS page, the next sectionname: Apartment # Display Name view: yes # Indicates if pagination is displayed at the beginning of the page entities: # Groups under the page – group.Entree – group.Salon – group.Cuisine – group.Bedroom1 – group.Bedroom2Entrance: name: The Entrance view: no icon: mdi: home-variant entities: – binary_sensor.sensor – light.level – sensor.alarm_typeLiving room: name: Salon view: no icon: mdi: sofa entities: – script.1529935670460 – script.1529935696908 – script.1529935786852 – script.1529935820515 – script.1529935881364 – script.1529935908101Cooked: name: Kitchen view: no icon: mdi: glass-tulip entities: – sensor.unknown_node_50_alarm_level – sensor.unknown_node_50_alarm_type – sensor.unknown_node_51_unknownRoom 1: name: Room1 view: no icon: mdi: pot entities: – script.1529935670460 – script.1529935696908 – sensor.unknown_node_51_alarm_typeBedroom2: name: Room2 view: no icon: mdi: hotel entities: – sensor.unknown_node_51_alarm_level – binary_sensor.unknown_node_50_sensor – script.1529935786852 – script.1529935820515ZWAVE_view: # This is the setting for the entire DEVICES page name: Z-WAVE DEVICES # Display Name view: yes # Indicates if pagination is displayed at the beginning of the page entities: – zwave.unknown_node_49 – zwave.unknown_node_50 – zwave.unknown_node_51 – zwave.unknown_node_52LIGHT_view: name: LIGHT view: yes entities: – script.1529936345615

Annex I.8 MATRIX Programming with Google Assistant

In order for Google Assistant software to access MATRIX Creator microphones, we must first install:# Add a repo and a keycurl https://apt.matrix.one/doc/apt-key.gpg | sudo apt-key add –echo “deb https://apt.matrix.one/raspbian $ (lsb_release -sc) main” | sudo tee /etc/apt/sources.list.d/matrixlabs.list# Update packages and installsudo apt-get updatesudo apt-get upgrade# Installation MATRIX Pacakagessudo apt install matrixio-creator-init# Rebootsudo rebootAfter the reboot go and install the kernel modules:# Installation MATRIX Packagessudo apt install matrixio-kernel-modules# Rebootsudo reboot# Then Configure the Google Assistant. In this step, you must configure all the necessary software to finally launch the wizard..Configure the Google Developer ProjectOpen the GOOGLE Console Actions Console. Click Add / Import Project. To create a new project, type a name in the Project Name box and click CREATE PROJECT.

Then choose a category or move on to choose later.

.Register device modelSelect the Device Registration tab (under ADVANCED OPTIONS) in the left navigation bar for your project. Click the REGISTER MODEL button.Fill in all the fields for your device. Select a device type, such as speaker.When done, click REGISTER MODEL. Download the file credentials.json on your PC and move them to raspberry pi in / home / pi / with MOBAXTERM. .Install the Google Assistant SDK and sample code. The Google Assistant SDK package contains all the code required for the Google Assistant to run on the device, including the sample code.Set up a new Python virtual environment$ sudo apt-get update

$ sudo apt-get install python3-dev python3-venv$ python3 -m venv env$ env / bin / python -m pip install –upgrade pip setuptools wheel$ source env / bin / activateInstall the system dependencies of the package:$ sudo apt-get install portaudio19-dev libffi-dev libssl-dev libmpg123-devInstall the latest version of the Python package in the virtual environment:$ python -m pip install –upgrade google-assistant-library$ python -m pip install –upgrade google-assistant-sdk [samples]Install or update the authorization tool:$ python -m pip install –upgrade google-auth-oauthlib [tool]Generate credentials to run the sample code and tools$ google-oauthlib-tool –scope https://www.googleapis.com/auth/assistant-sdk-prototype \ –scope https://www.googleapis.com/auth/gcm \ –save –headless –client-secrets /home/pi/credentials.jsonYou should see a URL displayed in the terminal:$ Please visit this URL to authorize this application: https: // …Copy the URL and paste it into a browser (this can be done on any machine). The page will ask you to sign in to your Google Account. Sign in to the Google account that created the developer project in the previous stepAfter approving the API authorization request, a code will appear in your browser, such as “4 / XXXX”. Copy and paste this code into the terminal:$ Enter the authorization code:If the authorization was successful, you will see a response similar to this one:$ credentials saved: /path/to/.config/google-oauthlib-tool/credentials.jsonGet code to use the Everloop$ git clone https://github.com/matrix-io/google-assistant-matrixio.git$ cd google-assistant-matrix /Start the Google Assistant!$ ~ / google-wizard-matrixio / google-matrixio-wizard-hotword –project_id matrix-creator-e92d7 –device_model_id matrix-creator-e92d7-matrix-creator-p3l4o6

Annex I.9 Installing Snips with Matrix creator

.Install Raspbian Stretch Install Matrix Kernel Modules: https://github.com/matrix-io/matrixio-kernel-modulescurl https://apt.matrix.one/doc/apt-key.gpg | sudo apt-key add –echo “deb https://apt.matrix.one/raspbian $ (lsb_release -sc) main” | sudo tee /etc/apt/sources.list.d/matrixlabs.listsudo apt-get updatesudo apt-get upgradesudo rebootsudo apt-get update – >> Not specified on the web page but otherwise gives an error:The following packages have unmet dependencies:matrixio-kernel-modules: Depends: matrixio-creator-init but it will not be installedDepends: dkms but it is not installableE: Unable to correct problems, you have held broken packages.sudo apt install matrixio-kernel-modulessudo rebootarecord -l:**** List of CAPTURE Hardware Devices ****card 1: Dummy [dummy], device 0: Dummy PCM [Dummy PCM]Subdevices: 8/8Subdevice # 0: subdevice # 0Subdevice # 1: subdevice # 1Subdevice # 2: subdevice # 2Subdevice # 3: subdevice # 3Subdevice # 4: subdevice # 4 Subdevice # 5: subdevice # 5 Subdevice # 6: subdevice # 6 Subdevice # 7: subdevice # 7card 2: SOUND [MATRIXIO SOUND], device 0: matrixio.mic.0 snd-soc-dummy-dai-0 []Subdevices: 1/1 Subdevice # 0: subdevice # 0.Install Snips: https://github.com/snipsco/snips-platform-documentation/wiki From: https://github.com/snipsco/snips-platform-documentation/wiki/1.-Setup-the-Snips-Voice-Platform Installing the Snips platform sudo apt-get update sudo apt-get install -y dirmngr sudo bash -c ‘echo’ deb https://raspbian.snips.ai/$(lsb_release -cs) stable main “> /etc/apt/sources.list.d/snips.list ‘ sudo apt-key adv –keyserver pgp.mit.edu –recv-keys D4F50CDCA10A2849 or sudo apt-key adv –keyserver pgp.surfnet.nl –recv-keys D4F50CDCA10A2849 sudo apt-get updatesudo apt-get install -y snips-platform-voice– skip the configuration part, should already be done.Install demo assistantapt-get install snips-platform-demo.Install Snips-Watchsudo apt-get install snips-watchCheck the status:sudo systemctl status “snips- *” = >> all should be runningStop Audio Serversudo systemctl stop snips-audio-serverOpen another terminal and syslog queuegsudo tail -f / var / log / syslogStart the audio server from the other terminalsudo systemctl start snips-audio-serverYou will see these errors: 

IMPORTANT TO FIX THE QUESTION:Delete DUMMY from matrix-mics.confsudo nano /etc/modules-load.d/matrix-mics.confcomment snd-dummy:snd_bcm2835# Snd-dummyrestart:Stop Audio Serversudo systemctl stop snips-audio-serverOpen another terminal and syslog tailsudo tail -f / var / log / syslogStart the audio server from the other terminalsudo systemctl start snips-audio-serverIn terminal systail, you should see: 

You have to configure /etc/snips.toml on raspberry pi to be able to run google assistant and snips in the same Matrix creator device.[Snips-audio-server]# frame = 256# bind = “0.0.0.0:26300”# mike = “Built-in Microphone”mike = “MATRIXIO SOUND: – (hw: 2,0)”

Annex II .1 Problems with taking out sensor parameters in HOME ASSISTANT

In the ZWAVE control panel of HA, it shows as below: “Sigma designs unknown: type = …”

 I use a Fibaro motion sensor motion sensor, a fibaro wall socket, a Zipato2 bulb, an AN145 on / off module to screw like ZWAVE devices. These devices can be added to HA but there is no content to configure for “node group associations” and “node configuration parameter”.

Regardless of which device I choose and they show the “Group”, “Configuration Setting” as the picture below, there is no data displayed. And the sensor was closed even though I clicked the B button three times.

 Take the example of Fibaro motion seonsor, after joining the network, it shows “node: 17 versions”, then I chose “binary_sensor.Sensor” in Entities of this node

 Then I chose “Alarm level” in the values ​​of the node. Then I would like to do the Node group association and the Config parameter, but after clicking the drop-down button, nothing changed. Always display “Config parameter”, “Group” if not.

In the zwave domain on the states page, after triple clicking on the sensor button b, it showed “initialization (Dynamic)”, the color of the sensor has changed to blue. But after a few seconds, the color turns off and displays “Sleeping”. Above, the “Alarm Level” button “Alarm type” always displays 0 and the sensor is closed.

Annex II .2 The problems of google assistant under Z-WAVE

After ordering~ / google-wizard-matrixio / google-matrixio-wizard-hotword –project_id your-dev-project-id –device_model_id your-model-idIt shows an errordevice_model_id: matrix-creator-e7-matrix-creator-p36device_id: BA335DBF6E9CE9C12B1AAD435015https://embeddedassistant.googleapis.com/v1alpha2/projects/matrix-creator-e92d7/devices/BA335DBF6E9CE9C12954B1AAD4350185 200EventType.ON_MUTED_CHANGEDEventType.ON_MEDIA_STATE_IDLETraceback (most recent call last): File “/ home / pi / google-wizard-matrixio / google-matrixio-wizard-hotword”, line 11, in sys.exit (main ()) File “/home/pi/google-assistant-matrixio/matrixio/hotword.py”, line 187, in main process_event (event, assistant.device_id) File “/home/pi/google-assistant-matrixio/matrixio/hotword.py”, line 68, in process_event set_everloop_color (0,0,0,10) # blue File “/home/pi/google-assistant-matrixio/matrixio/hotword.py”, line 53, in set_everloop_color with open (‘/ dev / matrixio_everloop’, ‘wb’) as bin_file:PermissionError: [Errno 13] Permission denied: ‘/ dev / matrixio_everloop’We type the commandls -l / dev / matrixio_ *the exit isls: can not access ‘/ dev / matrixio_ *’: No such file or directory

References

[1] C. Bailey and C. Sheehan, ‘Technology, older persons’ perspectives and the anthropological ethnographic lens’, ALTER – European Journal of Disability Research / Revue Européenne de Recherche sur le Handicap, vol. 3, no. 2, pp. 96–109, Apr. 2009.

[2] F. Boll and P. Brune, ‘Online Support for the Elderly – Why Service and Social Network Platforms should be Integrated’, Procedia Computer Science, vol. 98, pp. 395–400, 2016.

[3] E. P. W. A. Talboom-Kamp, N. A. Verdijk, M. J. Kasteleyn, M. E. Numans, and N. H. Chavannes, ‘From chronic disease management to person-centered eHealth; a review on the necessity for blended care’, Clinical eHealth, vol. 1, no. 1, pp. 3–7, Mar. 2018.

[4] L. Lahousse, H. Tiemeier, M. A. Ikram, and G. G. Brusselle, ‘Chronic obstructive pulmonary disease and cerebrovascular disease: A comprehensive review’, Respiratory Medicine, vol. 109, no. 11, pp. 1371–1380, Nov. 2015.

[5] D. Jain and V. Singh, ‘Feature selection and classification systems for chronic disease prediction: A review’, Egyptian Informatics Journal, Apr. 2018.

[6] E. P. W. A. Talboom-Kamp, N. A. Verdijk, M. J. Kasteleyn, M. E. Numans, and N. H. Chavannes, ‘From chronic disease management to person-centered eHealth; a review on the necessity for blended care’, Clinical eHealth, vol. 1, no. 1, pp. 3–7, Mar. 2018.

[7] M. Desmedt et al., ‘Economic Impact of Integrated Care Models for Patients with Chronic Diseases: A Systematic Review’, Value in Health, vol. 19, no. 6, pp. 892–902, Sep. 2016.

[8] C. Laurendeau, C. Chouaid, N. Roche, P. Terrioux, J. Gourmelen, and B. Detournay, ‘Prise en charge et coûts de la bronchopneumopathie chronique obstructive en France en 2011’, Revue des Maladies Respiratoires, vol. 32, no. 7, pp. 682–691, Sep. 2015.

[9] P. Skocir, P. Krivic, M. Tomeljak, M. Kusek, and G. Jezic, ‘Activity Detection in Smart Home Environment’, Procedia Computer Science, vol. 96, pp. 672–681, Jan. 2016.

[10] J. Schubert, S. Ghulam, and L. Prieto-González, ‘Integrated Care Concept using Smart Items and Cloud Infrastructure’, Procedia Computer Science, vol. 63, pp. 439–444, Jan. 2015.

[11] J. Sarasohn-Kahn, ‘A Role for Patients: The Argument for Self-Care’, American Journal of Preventive Medicine, vol. 44, no. 1, Supplement 1, pp. S16–S18, Jan. 2013.

[12] M. Sumit, A. Emad, N. Moein, and M.-T. Hamidreza, ‘Smart Homes for Elderly Healthcare—Recent Advances and Research Challenges’, Oct. 2017.

[13] A. Cohen and C. Leclercq, ‘Rapport coût/efficacité des traitements médicamenteux et interventionnels dans la fibrillation atriale’, Archives of Cardiovascular Diseases Supplements, vol. 8, no. 2, pp. 144–156, Feb. 2016.

[14] A. Berquin and J. Grisart, ‘To prescribe or not to prescribe… (in chronic pain… and elsewhere…)?’, Annals of Physical and Rehabilitation Medicine, vol. 54, no. 8, pp. 465–477, Nov. 2011.

[15] M. Labrunée, A. Pathak, M. Loscos, E. Coudeyre, J.-M. Casillas, and V. Gremeaux, ‘Therapeutic education in cardiovascular diseases: State of the art and perspectives’, Annals of Physical and Rehabilitation Medicine, vol. 55, no. 5, pp. 322–341, Jul. 2012.

[16] M. Labrunée, D. Antoine, B. Vergès, I. Robin, J.-M. Casillas, and V. Gremeaux, ‘Effects of a home-based rehabilitation program in obese type 2 diabetics’, Annals of Physical and Rehabilitation Medicine, vol. 55, no. 6, pp. 415–429, Sep. 2012.

[17] D. Gkouskos and J. Burgos, ‘I’m in! Towards participatory healthcare of elderly through IOT.’, Procedia Computer Science, vol. 113, pp. 647–652, Jan. 2017.

[18] J. M. Chapel, M. D. Ritchey, D. Zhang, and G. Wang, ‘Prevalence and Medical Costs of Chronic Diseases Among Adult Medicaid Beneficiaries’, American Journal of Preventive Medicine, vol. 53, no. 6, Supplement 2, pp. S143–S154, Dec. 2017.

[19] T. Tsukiyama, ‘In-home Health Monitoring System for Solitary Elderly’, Procedia Computer Science, vol. 63, pp. 229–235, Jan. 2015.

[20] V. Palanisamy and R. Thirunavukarasu, ‘Implications of big data analytics in developing healthcare frameworks – A review’, Journal of King Saud University – Computer and Information Sciences, Dec. 2017.

[21] S. Chakraborty, S. K. Ghosh, A. Jamthe, and D. P. Agrawal, ‘Detecting Mobility for Monitoring Patients with Parkinson’s Disease at Home using RSSI in a Wireless Sensor Network’, Procedia Computer Science, vol. 19, pp. 956–961, Jan. 2013.

[22] R. Goleva et al., ‘Performance Analysis of End-to-End Sensor-to-Cloud Personal Living Platform’, Procedia Computer Science, vol. 113, pp. 615–620, Jan. 2017.

[23] D. R. Kaufman et al., ‘Understanding workflow in telehealth video visits: Observations from the IDEATel project’, Journal of Biomedical Informatics, vol. 42, no. 4, pp. 581–592, Aug. 2009.

[24] C. Dourlens, ‘Les usages de la fragilité dans le champ de la santé’, ALTER – European Journal of Disability Research / Revue Européenne de Recherche sur le Handicap, vol. 2, no. 2, pp. 156–178, Apr. 2008.

[25] A. Hennion and P. Vidal-Naquet, ‘La contrainte est-elle compatible avec le care ? Le cas de l’aide et du soin à domicile’, ALTER – European Journal of Disability Research / Revue Européenne de Recherche sur le Handicap, vol. 9, no. 3, pp. 207–221, Jul. 2015.

[26] C. Moulart, H. Rienmeyer, and I. Tron, ‘Nutrition des personnes âgées brûlées en SSR’, Annals of Physical and Rehabilitation Medicine, vol. 57, p. e220, May 2014.

[27] A. Dinsenbacher, ‘Risque de chute et mesures de prévention des chutes pour des personnes âgées fragiles : l’exemple d’un hôpital de proximité’, Annals of Physical and Rehabilitation Medicine, vol. 57, p. e404, May 2014.

[28] L. de Waroquier-Leroy et al., ‘The Functional Reach Test: Strategies, performance and the influence of age’, Annals of Physical and Rehabilitation Medicine, vol. 57, no. 6, pp. 452–464, Aug. 2014.

[29] R. Hébert, M. Raîche, M.-F. Dubois, N. R. Gueye, and M. Tousignant, ‘Développement d’indicateurs pour valoriser des actions de prévention et de réadaptation de la perte d’autonomie des personnes âgées’, Revue d’Épidémiologie et de Santé Publique, vol. 60, no. 6, pp. 463–472, Dec. 2012.

[30] N. Barth, D. Hupin, F. Roche, T. Celarier, and B. Bongue, ‘La prescription de l’activité physique adaptée chez le sujet âgé : de l’intention à la réalité’, NPG Neurologie – Psychiatrie – Gériatrie, vol. 18, no. 105, pp. 155–161, Jun. 2018.

[31] Y. Huard, ‘Postures et gonalgie chronique chez les personnes âgées : contribution du traitement ostéopathique’, Annals of Physical and Rehabilitation Medicine, vol. 55, p. e101, Oct. 2012.

[32] M.Obach, P.Barralon, E.León, R&D projects related to all in tecnalia‘s health technologies unit, in:Hand book of Ambient Assisted Living,IOSPress,Amsterdam, pp.693–723,2012

[33] http://doc.openfing.org/CDP/Projets_CDP_PACA_190111/Gerhomelabs.pdf

[34] Les maisons «futées»:des Unités de Vie sécurisées pour Seniorsdes Unités de Vie sécurisées pour Seniors http://www.habiter-autrement.org/28_Homes/contributions-28/Projet-geriatric-maisons-fut%E9es-pour-seniors.pdf

[35] http://www.centich.fr/plateformes

[36] http://www.cmu.edu/qolt/

[37] B. Brumitt, B. Meyers, J. Krumm, A. Kern, and S. Shafer, ‘EasyLiving: Technologies for Intelligent Environments’, in Handheld and Ubiquitous Computing, Springer, Berlin, Heidelberg, 2000, pp. 12–29.

[38] H. Nakashima, H. Aghajan, and J. C. Augusto, Eds., Handbook of Ambient Intelligence and Smart Environments. Boston, MA: Springer US, 2010.

[39] http://ailab.eecs.wsu.edu/casas/

[40] https://www.cc.gatech.edu/fce/house/house.html

[41] ‘Ambient lighting assistance for an ageing population | Projects | FP6| European Commission’, CORDIS European Commission. [Online]. Available: https://cordis.europa.eu/project/rcn/80524_en.html

[42] A.Cesta,G.Cortellessa,M.V.Giuliani,etal.,The RoboCare Assistive Home Robot:Environment,Features and Evaluation.The RoboCare Technical Reports,Technical Report,National Research Council of Italy,2006

[43] http://www.aal-europe.eu/projects/icarer/

[44] K.Sakamura,Smart Living Environment: Ubiquitous Computing Approach Based on TRON Architecture,IOSPress,Amsterdam,2012

[45] T.Yamazaki,Theubiquitoushome,Int.J.SmartHome1(1) (2007) 17–22. http://refhub.elsevier.com/S2352-8648(15)00058-9/sbref15

[46] http://www.livinglabs.com.tw/en/index.php

[47] A. R. Salunke and A. N. Gaikwad, ‘Development of Unstructured Architecture for Voice and Data Services in Mobile Communication’, Procedia Computer Science, vol. 92, pp. 84–90, Jan. 2016.

[48] H. Chung, J. Park, and S. Lee, ‘Digital forensic approaches for Amazon Alexa ecosystem’, DigitalInvestigation, vol. 22, pp. S15–S25, Aug. 2017.

[49] O. L. Eyesan and S. R. Okuboyejo, ‘Design and Implementation of a Voice-based Medical Alert System for Medication Adherence’, Procedia Technology, vol. 9, pp. 1033–1040, Jan. 2013.

[50] A. Teixeira et al., ‘Speech-centric Multimodal Interaction for Easy-to-access Online Services – A Personal Life Assistant for the Elderly’, Procedia Computer Science, vol. 27, pp. 389–397, Jan. 2014.

[51] A. Hämäläinen, A. Teixeira, N. Almeida, H. Meinedo, T. Fegyó, and M. S. Dias, ‘Multilingual Speech Recognition for the Elderly: The AALFred Personal Life Assistant’, Procedia Computer Science, vol. 67, pp. 283–292, Jan. 2015.

[52] D. R. Large, L. Clark, A. Quandt, G. Burnett, and L. Skrypchuk, ‘Steering the conversation: A linguistic exploration of natural language interactions with a digital assistant during simulated driving’, Applied Ergonomics, vol. 63, pp. 53–61, Sep. 2017.

[53] R. Achkar, M. El-Halabi, E. Bassil, R. Fakhro, and M. Khalil, ‘Voice Identity Finder Using the Back Propagation Algorithm of an Artificial Neural Network’, Procedia Computer Science, vol. 95, pp. 245–252, Jan. 2016.

[54] K. J. Turner, ‘Analysing interactive voice services’, Computer Networks, vol. 45, no. 5, pp. 665–685, Aug. 2004.

[55] C. Krittanawong, H. Zhang, Z. Wang, M. Aydar, and T. Kitai, ‘Artificial Intelligence in Precision  Cardiovascular Medicine’, Journal of the American College of Cardiology, vol. 69, no. 21, pp. 2657–2664, May 2017.

[56] A. M. Peterson, D. P. Nau, J. A. Cramer, J. Benner, F. Gwadry-Sridhar, and M. Nichol, ‘A Checklist for Medication Compliance and Persistence Studies Using Retrospective Databases’, Value in Health, vol. 10, no. 1, pp. 3–12, Jan. 2007.

[57] J. Apolinário-Hagen, L. Fritsche, C. Bierhals, and C. Salewski, ‘Improving attitudes toward e-mental health services in the general population via psychoeducational information material: A randomized controlled trial’, Internet Interventions, vol. 12, pp. 141–149, Jun. 2018.

[58] V. Palanisamy and R. Thirunavukarasu, ‘Implications of big data analytics in developing healthcare frameworks – A review’, Journal of King Saud University – Computer and Information Sciences, Dec. 2017.

[59] P. J. Aiken and A. F. Rumbach, ‘Keeping the Voice Fit in the Group Fitness Industry: A Qualitative Study to Determine What Instructors Want in a Voice Education Program’, Journal of Voice, vol. 32, no. 2, pp. 256.e25-256.e34, Mar. 2018.

[60] A. Coucke et al., ‘Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces’, arXiv:1805.10190 [cs], May 2018.

[61] https://mycroft.ai

[62] https://www.allaboutcircuits.com/news/developing-open-source-voice-assistant-interview-mycroft-AI-Steve-Penrod/

[63] https://www.technologyreview.com/s/607892/an-open-source-and-cute-alternative-to-amazon-echo/

[64] https://www.neowin.net/news/mark-1-is-mycrofts-open-source-and-open-hardware-alternative-to-amazon-echo

[65] https://orbita.ai/2018/03/01/staywell-introduces-staywell-voice-leveraging-the-orbita-voice-platform-for-amazon-echo-and-chatbots-to-support-health-goals/

[66] https://www.mobihealthnews.com/content/orbitas-ai-voice-tech-supports-pillos-home-robot-and-other-digital-health-deals

[67] https://www.healthwise.org/press/healthwise-and-orbita-showcase-echo.aspx

[68] V. V. Garbhapu and S. Gopalan, ‘IoT Based Low Cost Single Sensor Node Remote Health Monitoring System’, Procedia Computer Science, vol. 113, pp. 408–415, 2017.

[69] N. F. B. I. Gulcharan, H. Daud, N. M. Nor, T. Ibrahim, and E. T. Nyamasvisva, ‘Limitation and Solution for Healthcare Network Using RFID Technology: A Review’, Procedia Technology, vol. 11, pp. 565–571, 2013.

[70] R. Challoo, A. Oladeinde, N. Yilmazer, S. Ozcelik, and L. Challoo, ‘An Overview and Assessment of Wireless Technologies and Co- existence of ZigBee, Bluetooth and Wi-Fi Devices’, Procedia Computer Science, vol. 12, pp. 386–391, 2012.

[71] http://kbase.x10.com/wiki/Main_Page

[72] I.-V. Sita and P. Dobra, ‘KNX Building Automations Interaction with City Resources Management System’, Procedia Technology, vol. 12, pp. 212–219, Jan. 2014.

[73] P. Zhou, X. Lei, and Z. Lv, ‘Study on Integrating BACnet with IPv6-based Wireless Sensor Networks’, Procedia Engineering, vol. 29, pp. 275–279, Jan. 2012.

[74] Z. Jiang and Y. Dai, ‘A decentralized, flat-structured building automation system’, Energy Procedia, vol. 122, pp. 68–73, Sep. 2017.

[75] https://fr.wikipedia.org/wiki/Courants_porteurs_en_ligne

[76] https://www.pedagogie.ac-aix-marseille.fr/upload/docs/application/pdf/2012-07/reseau_vdi.pdf

[77] A. K. Eidsvik, R. Karlsen, G. Blair, and P. Grace, ‘Interfacing remote transaction services using UPnP’, Journal of Computer and System Sciences, vol. 74, no. 2, pp. 158–169, Mar. 2008.

[78] https://www.wink.com/help/products/wink-hub/

[79] A. K. Eidsvik, R. Karlsen, G. Blair, and P. Grace, ‘Interfacing remote transaction services using UPnP’, Journal of Computer and System Sciences, vol. 74, no. 2, pp. 158–169, Mar. 2008.

[80] https://www.digitalilluminationinterface.org/

[81] http://www.homeplug.org/explore-homeplug/overview/

[82] A. Bakkali, J. Pelegri-Sebastia, T. Sogorb, A. Bou-Escriva, and A. Lyhyaoui, ‘Design and simulation of dual-band RF energy harvesting antenna for WSNs’, Energy Procedia, vol. 139, pp. 55–60, Dec. 2017.

[83] M. A. Choukou, B. Jarlot, A. Arfaoui, S. Bourelle, and R. Taiar, ‘Thermogrammes infrarouge : aide au diagnostic orthopédique’, Annals of Physical and Rehabilitation Medicine, vol. 56, p. e283, Oct. 2013.

[84] R. Muhendra, A. Rinaldi, M. Budiman, and Khairurrijal, ‘Development of WiFi Mesh Infrastructure for Internet of Things Applications’, Procedia Engineering, vol. 170, pp. 332–337, Jan. 2017.

[85] S. Tahir, S. T. Bakhsh, and A. H. Altalhi, ‘An Efficient Route Maintenance Protocol for Dynamic Bluetooth Networks’, Journal of King Saud University – Computer and Information Sciences, vol. 29, no. 4, pp. 449–461, Oct. 2017.

[86] S. Pathak, M. Kumar, A. Mohan, and B. Kumar, ‘Energy Optimization of ZigBee Based WBAN for Patient Monitoring’, Procedia Computer Science, vol. 70, pp. 414–420, Jan. 2015.

[87] C. W. Badenhop, S. R. Graham, B. W. Ramsey, B. E. Mullins, and L. O. Mailloux, ‘The Z-Wave routing protocol and its security implications’, Computers & Security, vol. 68, pp. 112–129, Jul. 2017.

[88] I. L. Ruiz and M. Á. Gómez-Nieto, ‘Combining of NFC, BLE and Physical Web Technologies for Objects Authentication on IoT Scenarios’, Procedia Computer Science, vol. 109, pp. 265–272, Jan. 2017.

[89] https://www.enocean.com/en/

[90] A. Ramesh, A. Naresh, and N. V. S. Rao, ‘Technique for Reduction of Intersymbol Interference in Ultra- Wideband System’, Procedia Technology, vol. 24, pp. 812–819, Jan. 2016.

[91] http://www.devlab.nl/projects/myrianed/

[92] http://www.dash7-alliance.org/

[93] https://www.insteon.com

[94] https://www.sigfox.com/en

[95] M. Bakula, P. Pelgrims, and R. Puers, ‘A Wireless Powering and Communication System for Implantable Devices Based on a Royer Oscillator with Radio and Near-field Communication Links’, Procedia Engineering, vol. 120, pp. 306–309, Jan. 2015.

[96] S. M. Hassan, R. Ibrahim, K. Bingi, T. D. Chung, and N. Saad, ‘Application of Wireless Technology for Control: A WirelessHART Perspective’, Procedia Computer Science, vol. 105, pp. 240–247, Jan. 2017.

[97] https://en.wikipedia.org/wiki/ANT_(network)

[98] http://www-users.cs.umn.edu/~tianhe/Papers/WEBee.pdf

[99] H. A. A. Al-Kashoash, F. Hassen, H. Kharrufa, and A. H. Kemp, ‘Analytical modelling of congestion for 6LoWPAN networks’, ICT Express, Dec. 2017.

[100]‘NeuRon/spl trade/ netform: a self-organizing wireless sensor network – IEEE Conference Publication’. [Online]. Available: https://ieeexplore.ieee.org/document/1043092/.

[101] https://www.matrix.one/products/creator

[102] https://www.home-assistant.io/

[103] https://ifttt.com/

[104] https://en.wikipedia.org/wiki/MIT_License

[105] https://www.openhab.org

[106] www.domoticz.com/

[107] http://vcbeat.net/ZTZhZjJhYjZmYjMxOGJiY2Y0NzRkYTVjM2Y1NDc5ZmI=

[108] https://orbita.ai/orbita-voice-exp-designer/

[109] https://matrix-io.github.io/matrix-documentation/#programming-layers

[110] J. Sin, C. Henderson, D. Spain, V. Cornelius, T. Chen, and S. Gillard, ‘eHealth interventions for family carers of people with long term illness: A promising approach?’, Clinical Psychology Review, Feb. 2018.

[111] D. Todolí-Ferrandis, J. Silvestre-Blanes, S. Santonja-Climent, V. Sempere-Paya, and J. Vera-Pérez, ‘Deploy&Forget wireless sensor networks for itinerant applications’, Computer Standards & Interfaces, vol. 56, pp. 27–40, Feb. 2018.

[112] H. Rahman and R. Rahmani, ‘Enabling distributed intelligence assisted Future Internet of Things Controller (FITC)’, Applied Computing and Informatics, vol. 14, no. 1, pp. 73–87, Jan. 2018.

[113] R. K. Barik et al., ‘Mist Data: Leveraging Mist Computing for Secure and Scalable Architecture for Smart and Connected Health’, Procedia Computer Science, vol. 125, pp. 647–653, Jan. 2018.

[114] J.-S. Oh and S.-B. Choi, ‘State of the art of medical devices featuring smart electro-rheological and magneto-rheological fluids’, Journal of King Saud University – Science, vol. 29, no. 4, pp. 390–400, Oct. 2017.

[115] C. Garcia-Perez et al., ‘Improving the efficiency and reliability of wearable based mobile eHealth applications’, Pervasive and Mobile Computing, vol. 40, pp. 674–691, Sep. 2017.

[116] C. W. Badenhop, S. R. Graham, B. W. Ramsey, B. E. Mullins, and L. O. Mailloux, ‘The Z-Wave routing protocol and its security implications’, Computers & Security, vol. 68, pp. 112–129, Jul. 2017.

[117] C. Wilson, T. Hargreaves, and R. Hauxwell-Baldwin, ‘Benefits and risks of smart home technologies’, Energy Policy, vol. 103, pp. 72–83, Apr. 2017.

[118] R. C. King, E. Villeneuve, R. J. White, R. S. Sherratt, W. Holderbaum, and W. S. Harwin, ‘Application of data fusion techniques and technologies for wearable health monitoring’, Medical Engineering & Physics, vol. 42, pp. 1–12, Apr. 2017.

[119] K. Morton et al., ‘Using digital interventions for self-management of chronic physical health conditions: A meta-ethnography review of published studies’, Patient Education and Counseling, vol. 100, no. 4, pp. 616–635, Apr. 2017.

[120] S. Sriramprakash, V. D. Prasanna, and O. V. R. Murthy, ‘Stress Detection in Working People’, Procedia Computer Science, vol. 115, pp. 359–366, Jan. 2017.

[121] A. Jurenoks and D. Jokić, ‘Sensor Network Information Flow Control Method with Static Coordinator within Internet of Things in Smart House Environment’, Procedia Computer Science, vol. 104, pp. 385–392, Jan. 2017.

[122] P. Brune and R. Rockmann, ‘Towards An Application Architecture For A Smart Online Service Network Platform For The Elderly’, Procedia Computer Science, vol. 113, pp. 442–447, 2017.

[123] B. Reeder and A. David, ‘Health at hand: A systematic review of smart watch uses for health and wellness’, Journal of Biomedical Informatics, vol. 63, pp. 269–276, Oct. 2016.

[124] L. Gabriele, ‘A Review of Systems and Technologies for Smart Homes and Smart Grids’, ResearchGate, no. Energies 9(348), pp. 1–33, May 2016.

[125] C. T. M. van Houwelingen, A. H. Moerman, R. G. A. Ettema, H. S. M. Kort, and O. ten Cate, ‘Competencies required for nursing telehealth activities: A Delphi-study’, Nurse Education Today, vol. 39, pp. 50–62, Apr. 2016.

[126] C. Palazzo et al., ‘Barriers to home-based exercise program adherence with chronic low back pain: Patient expectations regarding new technologies’, Annals of Physical and Rehabilitation Medicine, vol. 59, no. 2, pp. 107–113, Apr. 2016.

[127] E. Magill and J. Blum, ‘Exploring conflicts in rule-based sensor networks’, Pervasive and Mobile Computing, vol. 27, pp. 133–154, Apr. 2016.

[128] M. J. Coye, ‘Informatics: The Frontier of Innovation in Health and Healthcare’, Engineering, vol. 2, no. 1, pp. 37–39, Mar. 2016.

[129] E. Kilsdonk, L. W. Peute, R. J. Riezebos, L. C. Kremer, and M. W. M. Jaspers, ‘Uncovering healthcare practitioners’ information processing using the think-aloud method: From paper-based guideline to clinical decision support system’, International Journal of Medical Informatics, vol. 86, pp. 10–19, Feb. 2016.

[130] S. V. Zanjal and G. R. Talmale, ‘Medicine Reminder and Monitoring System for Secure Health Using IOT’, Procedia Computer Science, vol. 78, pp. 471–476, Jan. 2016.

[131] E. Pardo, D. Espes, and P. Le-Parc, ‘A Framework for Anomaly Diagnosis in Smart Homes Based on Ontology’, Procedia Computer Science, vol. 83, pp. 545–552, 2016.

[132] R. Negra, I. Jemili, and A. Belghith, ‘Wireless Body Area Networks: Applications and Technologies’, Procedia Computer Science, vol. 83, pp. 1274–1281, 2016.

[133] R. Li, B. Lu, and K. D. McDonald-Maier, ‘Cognitive assisted living ambient system: a survey’, Digital Communications and Networks, vol. 1, no. 4, pp. 229–252, Nov. 2015.

[134] B. M. C. Silva, J. J. P. C. Rodrigues, I. de la Torre Díez, M. López-Coronado, and K. Saleem, ‘Mobile-health: A review of current state in 2015’, Journal of Biomedical Informatics, vol. 56, pp. 265–272, Aug. 2015.

[135] M. Amiribesheli, A. Benmansour, and H. Bouchachia, ‘A review of smart homes in healthcare’, Journal of Ambient Intelligence and Humanized Computing, vol. 6, Mar. 2015.

[136] J. Puustjärvi and L. Puustjärvi, ‘The Role of Smart Data in Smart Home: Health Monitoring Case’, Procedia Computer Science, vol. 69, pp. 143–151, Jan. 2015.

[137] M. Fezari, R. Rasras, and I. M. M. E. Emary, ‘Ambulatory Health Monitoring System Using Wireless Sensors Node’, Procedia Computer Science, vol. 65, pp. 86–94, Jan. 2015.

[138] O. de M. Wagner, ‘Architecting Smart Home Environments for Healthcare: A Database-Centric Approach’, 2015.

[139] R.-H. Kim, ‘Cure Performance and Effectiveness of Portable Smart Healthcare Wear System Using Electro-conductive Textiles’, Procedia Manufacturing, vol. 3, pp. 542–549, 2015.

[140] S.-L. Chua and L. K. Foo, ‘Sensor Selection in Smart Homes’, Procedia Computer Science, vol. 69, pp. 116–124, 2015.

[141] A. Barberan-Garcia et al., ‘Effects and barriers to deployment of telehealth wellness programs for chronic patients across 3 European countries’, Respiratory Medicine, vol. 108, no. 4, pp. 628–637, Apr. 2014.

[142] H. C. Tung et al., ‘A Mobility Enabled Inpatient Monitoring System Using a ZigBee Medical Sensor Network’, Sensors (Basel), vol. 14, no. 2, pp. 2397–2416, Jan. 2014.

[143] N. Varga, L. Bokor, and A. Takács, ‘Context-aware IPv6 Flow Mobility for Multi-sensor Based Mobile Patient Monitoring and Tele-consultation’, Procedia Computer Science, vol. 40, pp. 222–229, Jan. 2014.

[144] A. Hussein, M. Adda, M. Atieh, and W. Fahs, ‘Smart Home Design for Disabled People based on Neural Networks’, Procedia Computer Science, vol. 37, pp. 117–126, Jan. 2014.

[145] J. Joe and G. Demiris, ‘Older adults and mobile phones for health: A review’, Journal of Biomedical Informatics, vol. 46, no. 5, pp. 947–954, Oct. 2013.

[146] N. Lasierra, A. Alesanco, S. Guillén, and J. García, ‘A three stage ontology-driven solution to provide personalized care to chronic patients at home’, Journal of Biomedical Informatics, vol. 46, no. 3, pp. 516–529, Jun. 2013.

[147] O. L. Eyesan and S. R. Okuboyejo, ‘Design and Implementation of a Voice-based Medical Alert System for Medication Adherence’, Procedia Technology, vol. 9, pp. 1033–1040, Jan. 2013.

[148] J. P. Amaro, F. J. T. E. Ferreira, R. Cortesão, and J. Landeck, ‘Powering Wireless Sensor Networks Nodes for Complex Protocols on Harvested Energy’, Procedia Technology, vol. 5, pp. 518–526, Jan. 2012.

[149] C. B. D. Kuncoro, Armansyah, N. H. Saad, A. Jaffar, C. Y. Low, and S. Kasolang, ‘Wireless e-Nose Sensor Node: State of the Art’, Procedia Engineering, vol. 41, pp. 1405–1411, 2012.

[150] L. D. Serbanati, F. L. Ricci, G. Mercurio, and A. Vasilateanu, ‘Steps towards a digital health ecosystem’, Journal of Biomedical Informatics, vol. 44, no. 4, pp. 621–636, Aug. 2011.

[151] C. Perumalla, J. Mak, N. Kee, and S. Matthews, ‘Integrating web applications to provide an effective distance online learning environment for students’, Procedia Computer Science, vol. 3, pp. 770–784, Jan. 2011.

[152] A. Queirós, M. Cerqueira, M. Santos, and N. P. Rocha, ‘Mobile Health to Support Ageing in Place: A Synoptic Overview’, Procedia Computer Science, vol. 121, pp. 206–211, Jan. 2017.

[153] H. Lucas, ‘New technology and illness self-management: Potential relevance for resource-poor populations in Asia’, Social Science & Medicine, vol. 145, pp. 145–153, Nov. 2015.

[154] G. Chetty, M. White, and F. Akther, ‘Smart Phone Based Data Mining for Human Activity Recognition’, Procedia Computer Science, vol. 46, pp. 1181–1187, Jan. 2015.

[155] A. R. Marra and M. B. Edmond, ‘New technologies to monitor healthcare worker hand hygiene’, Clinical Microbiology and Infection, vol. 20, no. 1, pp. 29–33, Jan. 2014

[156] http://aide-a-venir.com/