IoT based smart emergency response system (SERS) for monitoring vehicle, home and health status

doi:10.21203/rs.3.rs-4613881/v1

Download PDF

Research Article

IoT based smart emergency response system (SERS) for monitoring vehicle, home and health status

https://doi.org/10.21203/rs.3.rs-4613881/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

This paper presents an innovative Internet of Things (IoT)-based smart emergency response system that aims to improve safety and efficiency across vehicles, homes and healthcare settings. The proposed system combines advanced sensors, communication technologies, and intelligent algorithms to provide a comprehensive solution for real-time monitoring and emergency management. The vehicle device is equipped with temperature, gas, flame, GPS, piezo, and accelerometer sensor and the system can detect accidents and hazardous conditions, automatically notifying emergency services and providing precise location data. For home security, it incorporates Flame Sensor, humidity sensor, DHT11, MQ-2 gas sensor, MQ-4 Gas Sensor, and MQ-135 gas sensor, and intrusion detection to ensure immediate alerts and response to potential threats. In healthcare, the proposed health device is equipped with temperature, humidity, stress levels, heart rate, oxygen saturation, and fetal heart rate sensor which can monitor vital signs and detect anomalies, triggering alerts for medical emergencies and facilitating timely intervention. The system utilizes a cloud-based platform to collect, analyze, and disseminate data, enabling seamless coordination between emergency responders such as ambulance for health and accident emergency, fire brigade for fire emergency and affected individuals. We have also developed a MySQL-PHP-Laravel based live dashboard and emergency response system where both the user and relevant stakeholder will have the access which could mitigate the emergency situation in the quickest amount of time without human intervention. By leveraging IoT capabilities, the smart emergency response system offers significant improvements in response times, accuracy, and overall effectiveness of emergency management. This integrated approach not only enhances individual safety but also contributes to broader public safety goals, showcasing the potential of IoT technologies in creating smarter, safer environments. Further development and field testing are anticipated to refine the system's functionality and expand its application scope.

Emergency response system

vehicle device

home device

health device

accident detection

automatic notification

Internet of Things (IoT)

On a worldwide basis, the number of recorded traffic accidents and home emergencies has significantly increased in recent decades. According to the Global Status Report on Road Safety 2023, 1.19 million people—a lesser number—die in road accidents each year than in previous years. [1] Additionally, burns are caused by the breakdown of skin layers by hot liquids (scalds), solids (contact burns), or flames. Along with other factors like UV radiation, radioactivity, electricity, chemicals, and smoke inhalation, burns are a major global health concern, accounting for over 195,000 burn deaths annually. Over 95% of burn injuries that result in death occur in countries with poor and average incomes [2]. The amount of traffic and fire incidents in Bangladesh, one of the most populous nations in South Asia, is one of the most concerning issues. The Passenger Welfare Association of Bangladesh (PWAB) released a report stating that in 6,261 traffic crashes in Bangladesh in 2023, there were up to 7,902 fatalities and 10,372 injuries. Government statistics show that there were 24,074 fire incidents nationwide in last year compared to 19,642 in 2018, a 22.5 percent rise that resulted in a loss of Tk 330.41 crore [3]. Yet the actual rate of casualties will most likely be much greater. Between 1982 and 2000, there were 43% more accidents, but there were over 400% more fatalities during the same time span. Just before the festival festivities, it intensifies. In the last twenty-one years, there have been 84 thousand traffic accidents, resulting in 56 thousand fatalities and 63 thousand injuries. Many different factors contribute to road accidents. Additionally, the recorded information shows that between January 1, 1997, and December 31, 2018, there were approximately 250,000 fires nationally, based on data from the Fire Service and Civil Defense [4]. People's unconsciousness is one of the main causes of these incidents, and frequently, people's ignorance of the problem leads to the appropriate safeguards not being taken. Such events have much more disastrous aftermaths in terms of material and infrastructure losses, not to mention the staggering death toll [5]. Sadly, emergency services like fire, ambulances, and police have a longer reaction time in the event of an accident. This could be due to several factors, including high traffic, incomplete information about the accident, and a delay in notifying the proper authorities (such as by contacting an ambulance). Our project's development has been greatly influenced by this societal dilemma. We aim to address these challenges by proposing an innovative solution leveraging IoT technology. By integrating real-time data collection and analysis, live notification systems, and efficient emergency response mechanisms, this system seeks to mitigate the impact of road accidents and home emergencies, thereby improving public safety and saving lives. Also, research in this field holds immense promise for society as it endeavors to significantly reduce maternal mortality rates through swift and effective treatment management. This paper presents a comprehensive system to address the acute issues of road safety and emergency response in Bangladesh and abroad by thoroughly investigating the current state of facts and utilizing advanced technologies.

Innovation and the development of new technologies have facilitated the design and implementation of safety-promoting systems that are capable of identifying any possible risk and acting quickly to minimize its effects. The application of IoT technology is feasible in the areas of emergency response, accident prevention, and fire detection and response, among other areas. The current systematic literature review explores the prospects for integrating IoT into a comprehensive safety approach, as well as the recent and historical studies that are accessible in these domains.

Using TEMT6000 mild depth and MQ-02 fuel/smoke sensors along with Raspberry Pi 3 and Arduino Mega 2560 rev3 microcontrollers, Imteaj et al. (2017) invented an Internet of Things-primarily based fire alarming and authentication system workhouse. This novel technique indicates how current technology, such 360° servo motors and ArduCam MT9D111 video modules, can be included to improve fire detection and emergency response skills [6]. A most important turning point in hearth protection history was the 1812 installation of the first sprinkler device in history on the Theatre Royal on Drury Lane inside the United Kingdom. Automatic sprinkler structures had been evolved due to further tendencies by way of Philip W. Pratt in 1872 and Henry S. Parmalee in 1874. These systems are now required as protection devices in commercial and public buildings inside the United States and Canada[7].An automated fireplace detection device that forgoes conventional constant threshold values in favor of a statistical coloration model and binary heritage filters is presented in another study. This method's remarkable 85% detection rate without false positives, as shown by large-scale experiments, shows its effectiveness in fire detection when compared to traditional approaches.[8] Furthermore, a fire detection system that relies mostly on a wi-fi sensor network is good at identifying home fires early on, utilizing GSM output to reduce false alerts and ensuring energy-efficient sensor operation. This system uses simulation to show early hearth detection and highlights the need of immediate reporting to stop hazards to residential communities.[9] In the realm of accident mitigation, numerous measures, including airbags, ABS, and pre-collision technology, have been implemented to reduce the severity of accidents. However, there remains a dearth of comprehensive systems addressing post-accident emergency response. While existing studies propose smartphone-based accident detection systems, such as Faiz et al. (2017), our approach distinguishes itself by integrating IoT technology into a vehicle kit equipped with multiple sensors and a GPS module. This holistic system not only detects accidents but also provides detailed information to emergency responders, significantly improving post-accident response effectiveness. [10] Kodali and Ravi et al. [11] have developed a system that uses a Wi-Fi microcontroller and sensors such the smoke detection sensor (MQ-2), flame detection sensor, and flammable gas sensor (MQ-5) to detect fire in places of dwelling or work. Furthermore, integrated into the system is a GPS module. Thus, the alert message with the location is sent to the cloud service whenever there is a fire occurrence. All emergency services, including police and fire agencies, are linked in the cloud service. The fire department will take the required action after receiving the alert message.

On the other hand, a unique device was developed by L. K. Hema, Amalna, Athira Anil, and Kanya [12] to measure the heart rate of the fetus during pregnancy. The Fetal digital stethoscope sensor, which is placed on the pregnant woman's abdomen, is the central component of this device. A microprocessor processes the signals that the sensor records, allowing for precise determination of the fetal heart rate. Then, with the assistance of a GSM module, this data is texted to the appropriate mobile phone.

In a separate endeavor, George K. Endo, Ibukun Oluwayomi, Victor Alexandra, Yashodhan Athavale, and Sridhar Krishnan [13] introduced a system aimed at continuously monitoring crucial health parameters of pregnant women.This method makes use of a biomedical signal acquisition device when combined with sensors included into smartphones. Once the data has been processed and categorized, medical practitioners can use the insights gained to evaluate the health of pregnant women.

T. M. Kadarina and R. Priambodo [14] detailed an IoT-based totally real-time monitoring device for heart price and blood oxygen saturation tiers in both mothers and children. Employing a sensor module alongside Raspberry Pi, this machine enables the transmission of amassed information to the Things board IoT platform, making sure seamless tracking and management.

P. Poonkuzhlai, R. Aarthi, Yaazhini. V. M., Yuvashri. S, and Vidyalakshmi. G [15] delivered a complete system devoted to the non-stop monitoring of children's fitness parameters and their whereabouts for protection purposes. By monitoring metrics which include temperature, breathing rate, and heart charge, along with providing location monitoring, this gadget gives a clever solution for child tracking and monitoring.

The ability for transforming emergency reaction systems, twist of fate prevention, and fireplace detection is extensive when IoT technologies are integrated. Through the usage of superior sensors, microcontrollers, and networking technology, those systems provide previously unheard-of capacity for augmenting protection requirements in each industrial and home setting. The capacity to construct groups that are extra shrewd, steady, and resilient seems excellent so long as research on this region keeps growing. IoT-enabled fireplace detection systems are able to take proactive steps to save you from disastrous events by means of quick figuring out viable threats through real-time facts accumulating and analysis. IoT-based totally coincidence mitigation systems also decrease risks to human life and belongings harm by means of speedy detecting and responding to emergencies. Additionally, smooth integration of IoT devices with emergency response systems facilitates vocal communication and quick coordination, ensuring timely assistance in severe circumstances.

Similarly, the improvements incorporated into our device architecture are remarkable since they are unique features absent from previous research on health monitoring devices for children and expecting mothers. Our method includes all detectable feelings experienced throughout pregnancy in an attempt to lower the possibility of delivery problems. This comprehensive approach ensures that every measurable detail—from the fetus's movements to the mother's vital signs and symptoms—is meticulously monitored and assessed in real time. By providing pregnant mothers and doctors with practical knowledge and early warnings, our solution enhances maternal and fetal well-being by enabling them to take proactive measures. Furthermore, our device's smooth integration with the current healthcare system simplifies the provision of specific treatment and continuous monitoring, both of which enhance the health of kids and pregnant women.

Our system is intricately structured across four stages to ensure seamless operation and efficient response:

Stage 1: Networking and Sensing Layer

In this stage, a multitude of sensors including temperature, gas, flame, GPS, piezo, and accelerometer are strategically placed within the vehicle kit to gather real-time data. These sensors capture various parameters such as temperature, gas concentrations (including methane and butane), humidity, flame detection, acceleration, and impact. A microcontroller (Arduino Mega2560) and the GPRS module and sensors are integrated through PCB for data Collection.

Stage 2: Data Acquisition (DAQ) Systems

The microcontroller unit receives the collected data and uses it for storage, initial processing, and analog to digital conversion. Our method uses the Arduino Mega2560 directly, which simplifies the design and lowers costs, rather than the use of a separate DAQ board. Data transmission from sensors to the microcontroller occurs through I2C, serial, and analog protocols for converting to digital signals and sent to the Cloud Server for additional analysis.

Stage 3: Cloud Analytics

The GPRS module then sends the processed data to the server for comprehensive analysis. Our system is built using a dedicated physical server and is included in the cloud computing term of Infrastructure as a Service (IaaS). End users can use our mobile application to get real-time data updates. TCP protocol via the GPRS Module and MQTT protocol via the NodeMCU enable data transfer to the server. The relevant authorities are notified through our notification system if any accident occurs following the specific parameters which indicate emergency.

Stage 4: Identification and Assistance

The server evaluates the data it has received to see if the vehicle needs help. Based on the position of the car, the system determines the closest support provider, such as the Hospital or Fire Service, if assistance is considered necessary. The system matches sensor results to predefined threshold settings after identifying prospective threats. If these thresholds are exceeded, notifications are promptly sent to the mobile application users and relevant contacts. Through these four stages, our system ensures swift and effective response to critical situations, enhancing overall safety and security.

The diagram which is illustrated in figure 1 depicts a comprehensive emergency response system integrating data from vehicle, home, and health devices. After collecting all the data from different devices it transmitted all the collected data to a centralized cloud server which is known as a central database. The comprehensive device routes this data to different processing units for future prediction through NodeRed and Laravel. To identify the immediate emergencies here used real time data analysis whereas data is stored and forecast the future potential of emergencies based on the previous continuous stored data. Depending on the emergency the system notify the specific responder for the appropriate one including hospitals, police stations, fire brigades, and emergency contacts which ensures the rapid response to emergencies effectively and overall safety.

System working Mechanism:A process of data filtering procedures are used by this Internet of Things (IoT) system in the feedback and data storage operations. Different server scripts and programs enable each phase.

Server Data Reception:To capture data from both the home, vehicle and health devices, the initial step involves configuring the server accordingly (Figure 2.). We utilize a Virtual Private Server (VPS) as the central processing unit for this purpose. A versatile cross-platform data transfer software, NodeRed is connected to the server which operates a Linux-based operating system, CentOS 7. As an intermediary between several ways of data being sent and received, NodeRed operates. Regarding the wide range of connection protocols used for devices in homes as well as vehicles, NodeRed helps in efficiently handling streams of data. TCP and MQTT protocols for data receipt use specifically on the server. In order to enable MQTT data collecting, a MQTT broker is required. As the go-between, this broker is in charge of gathering data from linked devices and sending it to the users. The integration of Mosquitto and NodeRed enables the effective routing of data from all devices to the server for additional processing and analysis. The NodeRed flow of two devices are illustrated in figure 3.

Home Device: NodeMCU uses the MQTT protocol to send data to home devices. By using the MQTT Broker, "Mosquitto" accounts and passwords need to be created. In order to differentiate between multiple sources of data, information is received within "Topics" that have been determined. Data is sent by the Home Device NodeMCU with the subject name "home_kit," username, and password specified. When the MQTT broker at the server IP and port receives data related to this subject, it examines it and sends it to the NodeRED sketch for more investigation and implementation. Temperature, gas ppm readings, and flame detection information are all intended to be sent to the server via the Home device. In order to take preventative action and prevent disasters or address any unlikely scenarios that may have already happened, this makes it possible to identify potential fires or cooking gas leaks. The prototype home device illustrated into the figure 4a which was used for collecting data and took output.

Vehicle Device: The TCP protocol is used to receive data from the car device which is shown in figure 4b, simplifying the procedure and eliminating the need for a separate broker. Through a TCP port that has been assigned, in this example 1850, the data is sent publicly to a specified IP address. This data can be immediately received by NodeRED, which can then be transmitted to other services for further processing. Once the data is received, NodeRED can be configured to perform various tasks, such as converting it into an encoded message payload for additional evaluation.

Health Device: The health device also works similar to vehicle and home devices. The health device also collects data and transfers the data to the cloud server. The cloud server serves as a central database, storing and managing the collected data. Data is routed to various processing units using NodeRed and Laravel. We can visualize the data into our dashboard and LCD screen. The real time data processes a warning SMS also sent to the emergency contracts based on the intensity of the emergency which ensures the user's security. The usage data is stored in the cloud server for the future prediction of any emergency situations. We construct a prototype of the health device for proper data collection which is illustrated in the figure 4(c).

The information is collected and kept after it is taken. All data is put together as one string with commas. This data is kept in a MySQL database for more work. This is done with PHP, a language for servers. The PHP script divides the data by commas. It keeps each device's data in its own table. One table has all the data in order. Another table only has the newest data in one row. This is key for ongoing work and watching. This data logging gathers information for future updates. Then, another PHP script checks the latest data. It seems if emergency limits are met. If so, the user or an expert is needed. This script sends notifications to emergency responders for rapid assistance, or it warns users based on specific requirements. Each component of the data will be kept individually, after splitting the data, the script will save each device's information in two different tables. On the other hand, a single row containing only the current data will be used for monitoring and conditioning in the future. The purpose of the data log or storage table is to gather information for potential system updates. The database storing information and the calculation of the data for feedback or notification is illustrated in figure 5.

After the data has been saved, another PHP script is executed to make sure all the data is current and to define the parameter, which indicates whether a user query or expert assistance is needed. It sets data that needs to alert the user or notify the emergency responder for assistance, depending on the criteria.

Event Condition Action Rule (ECA) and Notification: Our system toggles the value of flag data in between 0 and 1 by using the event condition action rule in the script to compare the current parameters to previously recorded ones. Emergency responder services, including fire departments or hospitals, are prompted to take appropriate action in response to this flag adjustment. On the other hand, non-critical events cause mobile apps to send out notifications to users, warning them of minor catastrophes that don't call for medical attention.

Expanding our system to include health monitoring, we've conducted three fall detection test cases:

Monitoring a person in a sitting or standing position.
Detecting a fall where the person is unable to regain their feet or recover independently.
Identifying a fall where the person can get back up or recover without assistance.

Additionally, to maximize system performance, we have simulated a variety of emergency circumstances, such as:

The kitchen is on fire.
A kitchen gas leak
Gas data is not relevant in the case of the living room fire.
An engine fire in the car
The automobile is leaking gas.
severe impact on the car's side

Certain actions are triggered by each circumstance according to predetermined algorithm settings. Through focused testing and data flow analysis, our goal is to improve system responsiveness in various emergency scenarios.

We collect all of the sensor data for multiple hypothetical emergencies or scenarios for testing. After recording the sensor responses, we classified the severity level into three range values: low, minimal, and high risk. We will then alert the emergency responder or emergency contacts, depending on these emergency scenario severity levels.

5.1 Sensor Data Receive

Kitchen fires typically exhibit characteristics such as flames, elevated temperatures, and significant smoke production. The temperature, smoke, LPG, and flame detection levels in the kitchen and living room after different fire and gas leak incidents are shown in Table 1.Given the presence of fire, detecting LPG in such scenarios is less likely. Conversely, gas leakage incidents in the kitchen result in increased LPG levels in the surrounding environment. In the event of a living room fire caused by factors like electrical short circuits, LPG presence is minimal, with flames likely being of low intensity but temperatures elevated above normal. Smoke generation varies based on the severity of the fire.

Table 1: Detection levels in different incidents

Incidents	Flame	Temperature	LPG	Smoke
Fire in the Kitchen	150	53	3	1050
Gas leakage in the Kitchen	900	32	150	3
Fire in the living room	200	40	1	680
Fire in the living room	170	42	2	1260

Inorder to evaluate the severity of an issue, an emergency response system must precisely evaluate a variety of environmental elements. Flame, temperature, LPG, and smoke levels are used to trigger alerts or emergency actions, the below code sample illustrated. Each situation is assessed against predefined levels before notifications are sent to the user app or emergency response service.

# Define conditions to evaluate warning and emergency situations

warning = 1 if (flame < 500 or temperature > 49 or LPG > 500 or smoke > 600) else 0

emergency = 1 if (flame < 500 and smoke > 1000) or (smoke > 3000 and temperature > 50) else 0

# Test scenarios

scenarios = [

{"flame": 200, "temperature": 40, "LPG": 1, "smoke": 680},

{"flame": 170, "temperature": 42, "LPG": 2, "smoke": 1260}

]

In the previous code, the levels of flame, heat, gas, and smoke are assessed. The current level of each component is compared with predefined integer values and from that, it’s determined if an emergency or a warning is present. If the levels are greater than predefined values, the user app or emergency service is notified of a warning or an emergency. For instance, it will issue a warning if the flame level is excessive. In the event that there is smoke, gas, and extremely high temperatures, an emergency alert will be sent for assistance immediately. People can be kept safe by using the code to check these levels.

Different occurrences might lead to difficulties or malfunctions in vehicle events which we stated in table 2 and need quick action or remediation tactics. It is critical to understand the impacts of different circumstances to keep passengers safe and the vehicle in proper working condition, including engine fires, gas leaks, or crashes. Each case involves unique obstacles, which are usually presented in the form of sensor readings and critical system indicators changing.

Table 2: Detection levels in different incidents for the vehicle

Incidents	Flame	Smoke	CNG	ax	airbag	GPS	P1	P2	P3	P4	P5
Fire in the vehicle engine	50	630	93	0.3	0	23.770616,90.397149	100	115	103	94	137
Gas leakage in the vehicle	1023	2	543	0.12	0	23.770616,90.397149	103	105	104	103	98
Heavy collision on side of the vehicle	980	340	2	-6.5	1	23.770616,90.397149	102	115	863	917	137

# Define warning conditions

warning = 1 if (flame < 500 or smoke > 600 or CNG > 500 or ax < -10 or GPS == (0.0, 0.0) or any(p > 600 for p in [p1, p2, p3, p4, p5])) else 0

# Define emergency conditions

emergency = 1 if (airbag == 1 or (flame < 200 and smoke > 600) or (any(p > 600 for p in [p1, p2, p3, p4, p5]) and ax < -10)) else 0

The system then assesses whether any of the presented predefined thresholds meet the predefined warning or emergency condition while the sensor readings are available and the vehicle is in a normal condition. CNG levels are high, alarms are high flame, or alarms are excessive smoke, and a notification is issued via an app on the user’s phone. Notifications are also sent to emergency responders for quick engagement in scenarios where emergency conditions are recognized, such as airbag deployment or critical sensor readings, enabling rescue and fire suppression operations. No additional action is necessary if neither the warning nor the emergency conditions are met, guaranteeing effective resource allocation and reaction management.

Table 3: Summary table of health status

Parameter	Test Case 20-year-old pregnant woman
Temperature(°C) (T)	35.1[35]
Humidity (%) (H)	88.07 [85]
GSR (µS) (G)	323.53 [350]
Heart rate (bpm) (HR)	99[95]
SpO₂ (%) (S)	102.2 [97]
AD8232(mV) (A)	Fetal Heart rate detected: Yes
Health Status	T-Normal H-Normal G-Normal HR-Normal S-Normal

Pregnant women's health is evaluated by the health monitoring system using sensor data analysis. Normal and risk ranges are defined by reference values for blood oxygen saturation, heart rate, stress, temperature, and stress. For instance, a usual range of 32°C to 37°C was motivated by variations in temperature readings. Three thresholds for stress levels were identified: normal, low-risk (400-540µS), and high-risk (>540µS). Risk levels were also established by oxygen saturation and heart rate data. These thresholds are compiled and recorded data in table 3, which makes it easier for expectant patients to understand health data.

The system's responses for each of these test situations are shown in table 4.

Table 4: System Response for Test Cases

Test	Devices	Cases	Anticipated Reaction from the System	Findings
1	Home	Kitchen Flame	Notification sent to both app and emergency response service	Success
2		Kitchen Gas Leakage	Notification sent only to smartphone app	Success
3		Living Room Fire 1	Notification sent only to smartphone app	Success
4		Living Room Fire 2	Notification sent to both app and emergency response service	Success
5	Vehicle	Car Engine Fire	Notification sent to both app and emergency response service.	Success
6		Car Gas Leakage	Notification sent only to smartphone app	Success
7		Car Side Collision	Notification sent to both app and emergency response service	Success
8	Health	Woman Sitting or Standing	No trigger activated (No risk)	Success
9		Woman Fallen	Trigger 1 activated, Trigger 2 activated, Trigger 3 activated (High risk)	Success
10		Woman Fallen and Unable to Recover	Trigger 1 activated, Trigger 2 activated, Trigger 3 activated, Trigger 3 deactivated (No risk)	Success

All test scenarios provided successful results, accurately matching the system's expected responses. This shows that the system is capable of identifying and handling a range of emergency situations, as seen by the notifications that are sent to the emergency response agencies and the smartphone app as necessary. Additionally, to improve security and effectiveness, threshold values and notification mechanisms can be modified gradually.

5.2 Device Notification System Development

An Android device allows the user to search saved info instantly. For instance, when a user requests to access device data, the application will run a PHP script intended to display data in the "Current data" table. This script will ultimately provide the data from the specified rows and columns of the table. One data point will be displayed each second in the application. Every time the application uses the PHP script, it also verifies the data to see if there has been an accident or emergency. If this is the case, the user will get a notification on the app informing them of the occurrence. One app can monitor several devices because it is compatible with both home and car devices. The different interfaces of the application are shown in figure 6.

5.3 System for Emergency Responder:

When we use the word "Emergency responder," we mostly mean the fire department, ambulance service, hospitals, and civil defense. They are our system's important emergency responders. Since responders must react to user crises, the system's service is more effective in the areas where responders are registered.

Each Emergency Responder is equipped with a web dashboard from which they may access notifications about emergencies, such as building fires or local auto accidents. Their location will be displayed as their home location on a separate screen. An emergency flag will be present in the database when the previously described PHP scripts identify an emergency illustrated in figure 7 and figure 8.

The dashboard utilizes an API to check flag fields, ensuring prompt response to emergencies. If any device signals an emergency, the system retrieves its location and searches for nearby hospitals and fire services. Currently, five hospitals in Dhaka, including Dhaka Medical College Hospital and Apollo Hospital, are integrated into the system. During emergencies, the system compares the location with these hospitals using the API of the Google Maps Location, allowing the dashboard to display notifications visually and audibly. Once an emergency is identified, the dashboard provides location details for immediate assistance by the responder team. Also an SMS alert will be sent to the registered emergency contact number, one sms format is illustrated in the figure 8b.

5.4 Security and Privacy analysis

Our comprehensive system integrates three essential components: the home kit, vehicle kit, and health kit, each catering to a diverse user base. This framework systematically collects sensor data from several users and stores it in a MySQL database to enable continuous monitoring and conditioning procedures. Transmission of this to a cloud server is executed through both TCP and MQTT protocols, resulting in substantial data storage volumes on a daily basis. To fortify the security of this data, our system prioritizes encryption techniques and implements IP address access restrictions, addressing prior concerns regarding security and privacy. Notably, both the MySQL database and MQTT broker "Mosquitto" are password-protected, providing an additional layer of security for data sharing and storage. Furthermore, access to the system is restricted solely to specific IP addresses, further enhancing security protocols. When it comes to sharing data with emergency services, only pertinent information such as name, contact number, and live location is disclosed, prioritizing user privacy and confidentiality. Looking ahead, our future plans entail the incorporation of Google's firewall service to bolster user authentication measures. Notably, within the health kit, we leverage the ThingSpeak dashboard to enhance data visualization and analysis capabilities, further optimizing health monitoring functionalities for our users.

This paper presents an innovative Internet of Things (IoT) based Smart Emergency Response System (SERS) that integrates vehicle, home, and health monitoring into a comprehensive emergency management solution. Unlike traditional systems that address these domains separately, our approach unifies them, providing seamless real-time monitoring and automated emergency responses. The novelty of this system lies in its multi-domain integration, enabling it to handle diverse emergency scenarios with a single, coherent infrastructure. By utilizing advanced IoT technologies, cloud computing, and intelligence algorithms, MySQL-PHP-Laravel based live dashboard the proposed SERS not only enhances the accuracy and speed of emergency detection but also offers predictive analytics to foresee potential risks, thereby improving overall safety and response efficiency. The proposed comprehensive system has a number of advantages compared to existing literature which has been summarized in Table 5.

Table 5

Comparison of our proposed comprehensive system with existing literature
Services	Our Comprehensive System	Existing Works
Fire Detection and Response	Integrates IoT-based sensors and cloud computing for real-time fire detection, immediate response, and predictive analytics across vehicles, homes, and health systems.	Traditional fire detection systems frequently use stand-alone smoke detectors or sprinkler systems that lack integrated predictive analytics. For example, Imteaj et al. (2017) developed a fire alarm system using Raspberry Pi and Arduino [6], but it lacks predictive capabilities and integration across multiple domains.
Accident Detection and Response	Provides real-time accident detection with detailed data transmission (e.g., location, severity, health status) to emergency responders.	Faiz et al. (2017), uses smartphone-based detection [10], which may not be as reliable in severe accidents.These systems frequently have limited data transmission capabilities.
Health Monitoring	Monitors health factors (temperature, stress, heart rate, fetal heart rate), early warnings, integrates with healthcare.	Existing health monitoring solutions, such as the one proposed by L. K. Hema et al. (2020) [12], lacks integration with emergency response systems, limiting their effectiveness in critical situations.
Data Integration and Analytics	IoT-based health monitoring, data sent to customize dashboard, SMS alerts. Data processed by microcontroller, sent to cloud for analysis via MQTT.	Traditional systems generally do not leverage advanced analytics for proactive risk management.
User Accessibility and Interface	The user friendly interface,scalable design and real time data visualization makes the system noble compared to existing system	Many existing systems lack a user-centric design. Kodali and Yerroju's (2017) approach prioritizes technical functionalities [11] over user interface and scalability.

Limitations

The location of the service provider (hospital, police station, fire station) is crucial for our comprehensive system. In order to maximize the efficiency we need to register as many police, fire, and hospital stations as possible in our database. An agreement, or memorandum of understanding (MOU), must be signed in order to register a service. The main need to accomplish that is to persuade people to want to be a member of the system by selling its concept in a way that makes sense. Second, a separate request would need to be made to the government to sign a contract that permits their registration, since there are government-run hospitals in addition to privatized ones, as well as police and fire stations.
Data collecting requires hundreds, if not thousands, of trial runs before the kits are put on the market. Since this system is intended to identify incidents, a comparable simulated setting would need to be established in order to gather the best possible data. This would involve the deliberate happening of accidents in many situations, which is not entirely possible and could result in increased costs and resource usage.
Our system cannot notify the emergency contact person if the number we provide is unavailable for any reason, so it is essential that the number is always working.

Drawbacks

The primary disadvantage would be the service provider's unavailability. Because our system is only meant to alert one service provider (hospital, police, fire station) at a time, for example, in the event of an accident during rush hour at the alerted hospital, they would not be able to deploy an ambulance or offer immediate medical attention. The victim would therefore have to wait for assistance to be sent until the hospital is open again. The entire goal of speeding up reaction following an accident is defeated in this way.
Should the collision be too severe, the car or house kit may be destroyed and would need to be replaced afterwards. Alternatively, in a more severe and uncommon case, the kit might be destroyed before it can alert one of the service providers.
As a GSM module keeps the vehicle's navigation kit and cloud sensor connected continuously, traveling to a remote area without a network would disrupt the connection.
In the event that any of the kit's sensors become damaged or unusable, it will require human repair before it can begin collecting data once more.
An exponential rise in accident rates might occur in the exceptional event of a natural disaster, which would be detrimental to the system as a whole. In addition to the service providers being overworked, the massive volume of data increases the likelihood of a cloud server disaster.
Regarding the home kit, it is presumed that the residence where the kit is to be installed already possesses a basic fire alarm system, as is typical of most contemporary homes. Therefore, installing the house kit at a location without a basic sprinkler-equipped fire alarm system would be highly inappropriate.

Our project, like most others, has plenty of opportunities for advancement and progress.

We may further customize and increase the kits' exclusivity for the user by adding a fingerprint sensor. This gives the system a more personalized feel while also enhancing security and storing the user's unique information for easier identification in the event of an emergency.
In terms of vehicle kits, with a little bit of research, the more dangerous roads and intersections can be located in a specific spot where cameras can be installed. With rapid progress in image processing, photos from the crash scene can be taken in real time, verifying the data collected by the sensors. Photos captured in real time will give an improved understanding of the accident's location and scope. To have a better idea of the conditions of the occupants, cameras might be installed inside an automobile in addition to being fixed outside.
Moreover, a wearable health monitoring band that allows users to report more precise data about their physical status in real-time in both their daily lives and in the terrible event of an accident can be built and integrated to the entire system. In the event that a passenger wears a health-monitoring band, for instance, the vehicle kit has the ability to transmit information about their body temperature, blood pressure, and heart rate to the closest hospital for a more thorough examination of the circumstances.
It is possible to create an individual band with a focus on women's and children's safety. The band can be made so that, in the event of a potential threat of any kind, a notification is sent to the closest police station by just pressing a button.
With respect to the home kit, we may create a sprinkler system that will activate upon analysis of the kit's data and be utilized to extinguish small fires without overly informing the fire station.
We need to know the phone numbers of other family members or close relatives in case the emergency contact number is inactive and the threshold value is exceeded.
The Quectel GPS module must transmit the pregnant women's sensor data to the adjacent hospital in order for the user to get further help in the form of prompt action. This is extremely necessary since the mother may pass out in any unfavorable situation and be unable to respond on her own.

To conclude, in this study we have presented an smart IoT based comprehensive emergency response system that integrates vehicle,home and health monitoring systems in a unified system.The proposed system addresses the limitation of existing systems by providing automatic notifications, real time monitoring and predictive analytics across multiple domains.The incorporation of IoT technologies, cloud service and intelligence algorithm enables faster detection and rapid response with precedence speed and accuracy.The real time monitoring of several critical parameter for the combined system ensures system’s robustness in critical emergency situations, saving lives and minimizing damages. Additionally intelligence data visualization and analytics enhances proactive risk management forseeing potential emergencies before they occur, thereby improving preparedness and prevention strategies. The user-friendly interface, scalable design, big data analysis capabilities, and robust security protocol ensures the system is accessible, adaptable, scalable and capable of evolving future demand.

COMPETING INTERESTS

The author(s) declare no competing interests.

Author Contribution

Abu S. M. Mohsin conceptualized the idea, performed the analysis and prepared the draft, revised the main manuscript and supervised the team. Munyem Ahammad Muyeed performed the analysis and prepared the manuscript.

Acknowledgement

Authors would like to thank Sahriar, Zawata, Sakib, Sara, Fariha, Nabita, Anindita and Brac University, Bangladesh for the research support.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Global Status Report on Road Safety. 2023. https://www.who.int/publications-detail-redirect/9789240086517. Accessed 20 May 2024.
WHO EMRO | Burns | Health Topics. World Health Organization. - Regional Office for the Eastern Mediterranean, http://www.emro.who.int/health-topics/burns/index.html. Accessed 20 May 2024.
Saad M. Record Number of Fires Last Year: 24,074 Incidents, Tk 330cr Loss. Dly Star 10 Feb. 2020, https://www.thedailystar.net/frontpage/news/fire-incidents-spike-1865701.
Fire Incidents in Bangladesh Triple over 22 Years. Dhaka Tribune, 26. Feb. 2019, https://www.dhakatribune.com/bangladesh/nation/169805/fire-incidents-in-bangladesh-triple-over-22-years.
Aktar T, Fariha T. Preparedness for Fire Disaster. Asia Disaster Preparedness Center, https://www.adpc.net/igo/contents/iPrepare/iprepare-news_id.asp?ipid=231.Accessed 20 May 2024.
Imteaj A, Rahman T, Hossain MK, Alam MS, Rahat SA. An IoT based fire alarming and authentication system for workhouse using Raspberry Pi 3, 2017 International Conference on Electrical, Computer and Communication Engineering (ECCE), Cox's Bazar, Bangladesh, 2017, pp. 899–904, 10.1109/ECACE.2017.7913031.
Where Did Fire Sprinkler Systems Come From? ThoughtCo. https://www.thoughtco.com/fire-sprinkler-systems-4072210. Accessed 20 May 2024.
Cho B-H, Bae J-W, Jung S-H. Image Processing-Based Fire Detection System Using Statistic Color Model, 2008 International Conference on Advanced Language Processing and Web Information Technology, Dalian, China, 2008, pp. 245–250, 10.1109/ALPIT.2008.49.
Saeed F, Paul A, Rehman A, Hong WH, Seo H. IoT-Based Intelligent Modeling of Smart Home Environment for Fire Prevention and Safety. J Sens Actuator Networks. 2018;7(1):11. https://doi.org/10.3390/jsan7010011.
Faiz AB, Imteaj A, Chowdhury M, Engineering I. Smart vehicle accident detection and alarming system using a smartphone, (ICCIE), Rajshahi, Bangladesh, 2015, pp. 66–69, 10.1109/CCIE.2015.7399319.
Kodali RK, Yerroju S. IoT based smart emergency response system for fire hazards, in 2017 3rd International Conference on Applied and Theoretical Computing and Communication Technology (iCATccT), 2017, pp. 194–199: IEEE.
Hema LK, Amalna, Anil A, Kanya. (2020). Pregnant Women Health Monitoring System Using Embedded System. IOP Conference Series: Materials Science and Engineering. 993. 012078. 10.1088/1757-899X/993/1/012078.
Endo G, Oluwayomi I, Alexandra V, Athavale, Yashodhan, Krishnan, Sridhar. (2017). Technology for continuous long-term monitoring of pregnant women for safe childbirth. 6–10. 10.1109/IHTC.2017.8058200.
Kadarina T, Priambodo R. (2018). Monitoring heart rate and SpO 2 using Thingsboard IoT platform for mother and child preventive healthcare. IOP Conference Series: Materials Science and Engineering. 453. 012028. 10.1088/1757-899X/453/1/012028.
Poonkuzhlai P, Rameshkumar A. Child Monitoring and Safety System Using Wsn and Iot Technology. Annals Romanian Soc Cell Biology. 2021;25:10839–47.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
07 Aug, 2024
Reviews received at journal
05 Aug, 2024
Reviewers agreed at journal
31 Jul, 2024
Reviews received at journal
10 Jul, 2024
Reviewers agreed at journal
03 Jul, 2024
Reviewers agreed at journal
03 Jul, 2024
Reviewers invited by journal
03 Jul, 2024
Editor assigned by journal
01 Jul, 2024
Submission checks completed at journal
01 Jul, 2024
First submitted to journal
20 Jun, 2024

You are reading this latest preprint version

IoT based smart emergency response system (SERS) for monitoring vehicle, home and health status

Status:

Version 1

Abstract

Figures

1. INTRODUCTION

2. RELATED WORK

3. DESIGN AND DEVELOPMENT

4. DATA FLOW ANALYSIS

5. RESULTS AND DISCUSSION

6. NOVELTY AND CONTRIBUTION OF THIS STUDY

7. LIMITATIONS & DRAWBACKS

8. FUTURE WORK & POSSIBILITIES

9. CONCLUSION

Declarations

COMPETING INTERESTS

Author Contribution

Acknowledgement

DATA AVAILABILITY STATEMENT

References

Additional Declarations

Status:

Version 1