A Mobile IoT-based Elderly Monitoring System for Senior Safety

Traditional elderly monitoring systems may misdiagnose and treat senior citizens due to inaccurate results, resulting in higher care costs and poorer health outcomes. Thus, the Internet of Things (IoT) was introduced to provide accurate, real-time monitoring data to improve overall performance. The IoT health system connects devices wirelessly to collect and analyze health data, monitor vital signs and physical activity, and provide real-time insights to improve health management. This paper aims to monitor real-time conditions as well as ensure the safety of senior citizens by developing a low-cost wearable prototype device to measure heart pulses, detect falls, and determine their actual location in an indoor space. The data is uploaded to IoT platforms like Blynk, Firebase, and Google Assistant, providing frequent updates on the elderly's health status and conditions and sending emergency messages, if necessary, to an Android application. A home control system is also developed to control the home appliances using mobile phones or voice control. A demonstration has been conducted to showcase the operation and functionality of the prototype. The proposed system has the capability to simplify the daily routines of the elderly while also granting caregivers greater control over their health and well-being.

Elderly monitoring; Fall detection; Indoor positioning; Internet of Things

System Architecture

2.1. Hardware Description

       The overall architecture is designed to remotely monitor the elderly’s health conditions and assist the guardian in managing their conditions easily using mobile phones. It consists of Sensing, Monitoring, and Home Control Devices, as explained in the next section.

2.1.1. Sensing device

    The first prototype, as shown in Figure 1(a), is equipped with a Wemos D1 mini microcontroller. The device is frequently updated on the elderly person's state by using the fall detection sensor to determine if there is a fall or not, the heart rate sensor to determine the heart rate pulses, the internal Wi-Fi module to determine the location of the elderly person inside the home, and the push button to send an emergency message. The Wemos D1 is chosen because of its ability to provide more power efficiency in a small package and its low cost. The system uses only a 3.7 V rechargeable Lithium Polymer (LiPo) battery.

      The fall detection sensor detects the fall by comparing the threshold value with the acceleration magnitude. If there is a change, it will take action. Both “Blynk” and “IFTTT”, which are “IoT platform” and “If This Then That” services, respectively, are making the action process (Blnyk, 2023; IFTTT, 2021). Blynk is an online platform that shares all the data collected from the sensors and stores it in the Firebase database (Firebase, 2023). IFTTT is a service that provides multiple options for taking action if something is triggered. The advantage of using IFTTT is that you can send a warning message to a specific number, send an email, etc. For this sensor, the output is chosen to be an emergency email message to the guardian. Meanwhile, the heart rate sensor senses the pulse of the heart whenever an elderly person holds it. The LED is used to blink simultaneously with the pulses. The heart rate sensor readings are shown in the mobile application, which will be discussed later. Lastly, the push button is programmed as an emergency button to send an emergency message if pressed. This method is done through the use of the “IFTTT” service.

Figure 1 (a) Sensing device, (b) Monitoring device, and (c) Home control device

      By using the internal Wi-Fi module, the location of the elderly person can be detected via the Received Signal Strength Indicator (RSSI) measurement between the fixed-place router in the house and the Wi-Fi module inside the Wemos D1 mini (Kawecki, Hausman, and Korbel, 2022). After the location is determined, it is frequently updated in Firebase and represented in the mobile Android application via Blynk. The power consumption is low compared to systems mentioned in the literature review because of the use of a small number of communication modules, like Wi-Fi and Bluetooth modules, and because no logic gates are used to power some sensors that depend on a specific turn-on voltage.

2.1.2. Monitoring device

       The second prototype, as shown in Figure 1(b), describes the ability to send and receive data under the guardian’s control. This device is equipped with a NodeMCU (ESP8266) microcontroller. It has two buttons to set a quick reminder on Google Calendar to take medicine (1) and medicine (2), respectively. The same methodology is applied here to send a message using the IFTTT service to make an event on Google Calendar if the button is pressed. Also, this device will have an implementation of a Liquid Crystal Display (LCD) with a size of 16 x 2 to update data (heart rate reading, elderly person's current location, and status of home appliances) from the Firebase via the IoT platform regarding the status of the elderly person and the interaction with the guardian if the button is pressed.

2.1.3.  Home control device

       Voice control is utilized to operate home appliances like light bulbs. Google Assistant, a mobile virtual assistant software program developed by Google, is programmed by the IFTTT service to execute specific voice commands for controlling home appliances. There is also an implementation of a virtual button on the Blynk mobile application to control the home appliances manually. The connection of the prototype is shown in Figure 1(c), in which the NodeMCU is used with a one-channel relay. Relay is the link between home appliances and the microcontroller. These two methods of controlling the home appliances, either by voice control (Google Assistant) or manual control (Blynk), are provided to the elderly to ensure there will be no difficulties in controlling the appliances. Firebase is used to update the status of the light bulb.

2.2. Software Description

        Mobile phones play an important role in this research. It is used to monitor the elderly person's health by showing the readings of the sensors in the Blynk application, as shown in Figure 2. It first uses the light bulb virtual button to control the light in the elderly’s home. Second, the location detection label is frequently updated with the home location changes of the elderly. Finally, the heart rate label shows the pulse readings of the heart rate sensor whenever it is handled. Moreover, the phone has a voice control feature to control the home appliances using Google Assistant. It controls the light status in the home by saying specific words, as shown in Figure 3. The voice control will also update the light virtual button status in Blynk using the IFTTT service. The Firebase is the prime link among all the devices in this research. The Sensing Device depends on it to send data to be viewed by the Monitoring Device. The Home Control Device also depends on Firebase to control the data. 

Figure 2 Blynk application                        Figure 3 Google Assistant service

2.3. Algorithm and Flow Charts

     The system's algorithm operates within the interactions among the devices mentioned earlier and the IoT platforms. It provides a systematic process flow for collecting health data and monitoring the conditions of the elderly, including fall detection and location tracking, all while enabling remote monitoring by caregivers. The devices serve as the front end of the system, while the IoT platforms are used as the back end to process, store, and manage the information received, as shown in the following flowcharts.

        The flowchart of the first prototype is shown in Figure 4. The system will check whether the button is pressed or not. If it is pressed, it will request a web request from IFTTT to send an emergency email to the guardian. The second function is to measure the heart rate pulses if an elderly person holds the sensor. The LED is blinking in synchrony with pulses. Then, the measurement data will be uploaded continuously to Blynk and Firebase. The third function is to detect a fall. After detection, it will send a web request to IFTTT to send another email to the guardian, warning him of fall detection.   The flowchart of the second prototype, as shown in Figure 5, starts by reading the data uploaded and saved previously on the Blynk. These data are displayed on the LCD in a continuous loop. The data types shown are heart rate pulses, location detection, and home appliance status. Then, the system detects whether one of the buttons is pressed or not. If it is pressed, a web request will be sent to IFTTT to make a medicine reminder on the Google calendar. There are two reminders to set for each of the buttons, respectively. Both buttons have the same sequence for applying the same function, except for different message outputs.

Figure 5 Monitoring device flowchart

       The flowchart of the third prototype, as shown in Figure 6, has two functions. They are either voice-controlled or manually controlled by the mobile phone. Voice control starts by saying, “Hey, Google,” then controlling the light by saying, “Activate light ON” or “Activate Light OFF.” Each method sends a web request to IFTTT to control the light’s status. After sending it to IFTTT, the light’s status is updated on Firebase and Blynk. The stored data is updated in Blynk to be controlled by the virtual button. Manual control is done by using the Blynk application with a virtual button to manually control the light by pressing the button to turn it on or off. Lastly, the manual control updates the data on the Firebase.

Figure 6 Home control device flowchart

     In this section, a demonstration of the developed prototypes, as shown in Figure 7, will be performed. The interactions between these devices will also be explained.

Figure 7 Developed prototype

3.1.  Sensing Device

        The first prototype has multiple sensors to check the health of the elderly person. The software is coded to read data from each sensor and update it simultaneously on both the Blynk and Firebase IoT platforms. The hardware is designed to fit all the sensors in one small prototype device, and it is supported with the battery shield to have a stand-alone system with a rechargeable small battery, which will not affect the weight. The output of the heart rate sensor is blinking the LED synchronously with the heart pulses, saving the latest readings in the Firebase platform to be represented on the monitor device later on and uploading the data to the Blynk IoT platform to be presented in the mobile application. To validate the heart rate sensor's measurement values, it has been tested along with the Mi Band 6 smart watch. Table 2 compares the readings from the heart rate sensor and the smart watch Mi Band 6. These readings are calculated after measuring at different times and in different situations. It’s important to mention that the heart rate sensor implemented inside the smartwatch is more advanced than the one used in our prototype system and costs much more than our sensor, but the readings are mostly the same, with a maximum relative difference of less than 3%.

Table 2 Comparison between the heart rate sensor and MI Band 6 smartwatch

       The MPU-6050 fall detection sensor accurately detects falls and triggers a rapid response: sending an emergency email to the guardian for assistance, as depicted in Figure 8. This email-sending function is facilitated through the IFTTT service, as it is no longer supported by other IoT platforms. It has the flexibility to modify the message or even the action itself after programming the microcontroller and has a fast response if the action is triggered, like falling. A quick response is necessary to help the elderly as soon as possible. So, the time spent sending the email is only 1 second. On the other hand, the emergency button is provided to contact the guardian via email and ask for help with a single press, which results in a fast response without any delay or limitations. The email notification is shown in Figure 9.

Figure 8 Email notification on fall detection               Figure 9 Emergency notification

      With the aid of the RSSI technique and Wi-Fi, the location of the elderly person inside the home can be easily determined, and the data will be uploaded to Firebase and Blynk simultaneously. If it is not repeated, it will save the microcontroller's memory and prevent the system from failing and stopping. Table 3 shows the ranges of the RSSI preserved for the specific home location. These data are measured by calculating the distance between the home router and the internal Wi-Fi module inside the Wemos. Since the prototype is placed in the living room, the nearest detected location is the living room space that has the highest RSSI range value. Where else the furthest place is room, which has the lowest range RSSI value. The internal Wi-Fi module also helps decrease the prototype device's size.

Table 3 RSSI definition for specific home location

3.2.  Monitoring Device

       The second prototype is developed for internal home use only. It has two functions: presenting the data saved from Firebase and setting a medicine reminder on Google Calendar by using the IFTTT service. The LCD shows the data in a looped sequence of location, heart rate, and light status all the time, as shown in Figure 10. These data are updated simultaneously with the latest values in the Firebase. The loop design of presenting the data gives the guardian person the smoothness of monitoring the health of the elderly person without any manual control on the screen. On the other hand, the two reminder buttons result in making a reminder on the Google Calendar, as shown in Figure 11. The first and second buttons send the first and second medicines’ names, respectively. The result will be shown on the computer or mobile phone to remind the elderly to take a specific medicine at a particular time. The medicine's time and name can be modified from the mobile device or the computer by using the IFTTT service at any time. the computer by using the IFTTT service at any time.

Figure 10 Data loop on LCD                                       Figure 11 Reminders on Google Calendar

3.3.  Home Control Device

     The third prototype is designed to control home appliances with either voice control or manual control from the mobile phone. The connection between the mobile phone and the microcontroller is made using Blynk. The connection between the Google Assistant on the mobile phone and the microcontroller is made using the IFTTT service. The connection between these services and IoT platforms is shown in Table 4. It goes for the sequence started by either Google Assistant (voice control) or a mobile application (manual control). In Google Assistant, it is controlled by saying, “Hey Google, Activate Light ON” or “Hey Google, Activate Light OFF” in order to send to the microcontroller either “0” or “1”, respectively. The manual control in the mobile application is done by a virtual button in the mobile app. This virtual button sends “0” or “1” if the button is switched to be turned “ON” or “OFF,” respectively. Then, all of these states use the relay to control the light bulb. The type of relay used is an “active low” relay, which means the relay will switch on the light bulb if it receives “0” and vice versa. At the last stage, the light status will be updated in Firebase and shown on the Monitor Device.

Table 4 Home appliances control sequence

3.4.  Comparisons with Existing Systems

        After showcasing the functionality of the developed prototypes, it can be concluded that the system is practically feasible to implement in the real world. Compared with some of the previous systems, as shown in Table 1, the developed system has a lower cost and a smaller size. The choice of using the NodeMCU ESP8266 and Wemos D1 mini rather than Arduino successfully reduces the size of the hardware. Furthermore, the Sensor Device is wearable, which is portable and convenient for the elderly. All the sensors and units are also tested and validated with high accuracy, as are the existing works. The mobile and web applications are user-friendly and yet add useful functions such as indoor positioning, voice control, etc., which are not introduced in the other existing systems.

       The main contribution of this paper is successfully developing a health monitoring system for elderly people using IoT technology and mobile applications. The prototype is low-cost and low-power, and it can be easily implemented in real-life applications such as medical systems. The system uses an Arduino microcontroller incorporated with multiple sensors to measure heart pulses, detect falls, and provide location data through an Android application. The application also controls home appliances, sends emergency emails, and provides medicine reminders. The wearable device is placed on the hand to easily access emergency buttons and simulate a real smartwatch. The significant of this work is that the system successfully provides real-time health data to carers and improves the daily lives of the elderly. The advantages of the system developed for the health monitoring industry, such as hospitals, are that it saves doctors time and effort in monitoring their patients’ health status. In emergencies, fast contact between the patient and the doctor is also achieved. As the prototype device delivers real-time data, the sensors’ readings are updated second by second. Also, patients will not face difficulties wearing the prototype device because of its small size. The limitations of this prototype device can be found in the packaging part. For future development, sensors can be covered; since this device is designed for the elderly, the risks of getting these sensors damaged are increasing. The mobile application can be improved by adding a user profile for each patient. It will keep their information secure and only accessible to authorized doctors. That will also make it easier for a medical specialist to monitor multiple patients simultaneously.

    This research project is fully sponsored by Internal Research Fund (MMUI/220008), Multimedia University.

Abdullah, M.I., Raya, L., Norazman, M.A.A., Suprihadi, U., 2022. Covid-19 Patient Health Monitoring System using Internet of Things (IoT). In: Institute of Electrical and Electronics Engineers (IEEE) 13^th Control and System Graduate Research Colloquium (ICSGRC), pp. 155–158

Bharathi, M., Rani, D.L., Padmaja, N., Dharani, M., 2022. A Health Monitoring System Based on Internet of Things (IoT) for Persons in Quarantine. Journal of Algebraic Statistics, Volume 13(3), pp. 387–392

Blynk, 2023. Introduction - Blynk documentation. Available Online at https://docs.blynk.io/en, Accessed on January 15, 2023

Chaudhary, R., Jindal, A., Aujla, G.S, Kumar, N., Das, A.K., Saxena, N., 2018. Lattice-Based Secure Cryptosystem for Smart Healthcare (LSCSH): Lattice-based Secure Cryptosystem for Smart Healthcare in Smart Cities Environment. Institute of Electrical and Electronics Engineers (IEEE) Communications Magazine, Volume 56(4), pp. 24–32

Cheng, B.J., Jamil, M.M.A., Ambar, R., Wahab, M.H.A., Ma’radzi, A.A., 2020. Elderly Care Monitoring System with Internet of Things (IoT) Application. In: Recent Advances in Intelligent Information Systems and Applied Mathematics, pp. 525–537

Chin, C.G., Jian, T.J., Ee, L.I., Leong, P.W., 2022. Internet of Things (IoT)-based Indoor and Outdoor Self-quarantine System for COVID-19 Patients. International Journal of Technology, Volume 13(6), pp. 1231–1240

Firebase, 2023. Firebase Application Programming Interface (API) Reference. Available Online at: https://firebase.google.com/docs/reference, Accessed on January 15, 2023

If This Then That (IFTTT), 2011. Documentation-Build Your Integration. Available Online at: https://ifttt.com/docs, Accessed on January 15, 2023

Irianto, B.G.,  Maghfiroh, A.M., Sofie, M., Kholiq, A., 2023. Baby Incubator with Overshoot Reduction System using Proportional Integral Derivative (PID) Control Equipped with Heart Rate Monitoring Based on Internet of Things. International Journal of Technology, Volume 14(4), pp. 811–822

Junaid, S.B., Imam, A.A., Balogun, A.O., De-Silva, L.C., Surakat, Y.A., Kumar, G., Abdulkarim, M., Shuaibu, A.N., Garba, A., Sahalu, Y., Mohammed, A., Mohammed, T.Y., Abdulkadir, B.A., Abba, A.A., Kakumi, N.A.I., Mahamad, S., 2022. Recent Advancements in Emerging Technologies for Healthcare Management Systems: A Survey. Healthcare, Volume 10(10), pp. 1940

Kawecki, R., Hausman, S., Korbel, P., 2022. Performance of Fingerprinting-based Indoor Positioning with Measured and Aimulated Receiver Signal Strength Indicator (RSSI) Reference Maps. Remote Sensing, Volume 14(9), pp. 1992

Saleem, K., Bajwa, I.S., Sarwar, N., Anwar, W., Ashraf, A, 2020. Internet of Things (IoT) Healthcare: Design of Smart and Cost-effective Sleep Quality Monitoring System. Journal of Sensors, Volume 2020, pp. 1–17

Sangeethalakshmi, K., Preethi-Angel, S., Preethi, U., Pavithra, S., Shanmuga-Priya, V., 2021. Patient Health Monitoring System using Internet of Things (IoT). Materials Today: Proceedings, Volume 80, pp. 2228–2231

Tamilselvi, V., Sribalaji, S., Vigneshwaran, P., Vinu, P., GeethaRamani, J., 2020. Internet of Things (IoT) Based Health Monitoring System. In: 2020 6^th International conference on advanced computing and communication systems (ICACCS), pp. 386–389

Tastan, M., 2018. Internet of Things (IoT) Based Wearable Smart Health Monitoring System. Celal Bayar University Journal of Science, Volume 14(3), pp. 343–350

Whulanza, Y., Kusrini, E., 2023. Defining Healthy City and Its Influence on Urban Well-being. International Journal of Technology, Volume 14(5), pp. 948–953