DESIGN OF WIRELESS SENSOR NETWORK (WSN) BASED LIFE SAVE SYSTEM FOR SCAVENGING & FIRE FIGHTING OPERATIONS

Transcription

1 Volume 118 No , ISSN: (printed version); ISSN: (on-line version) url: doi: /ijpam.v118i11.61 ijpam.eu DESIGN OF WIRELESS SENSOR NETWORK (WSN) BASED LIFE SAVE SYSTEM FOR SCAVENGING & FIRE FIGHTING OPERATIONS 1 Wisemin Lins, 2 T.R.Premila, 3 N.Janaki Assistant professor Dept. of EEE, Vels Institute of Science, Technology and Advanced Studies, Chennai, India 1 wisemin.se@velsuniv.ac.in Abstract: Yearly there are 250+ deaths are occurring due to scavenging and fire fighting operations. Most of the deaths are due to presence of poisonous gases being present in the scavenging digs and fire fighting areas. If we able to inform fire fighter or scavenging worker about the presence of gases and hazardous situation, we can able to save their lives. Wireless sensor networks (WSN) have been employed to collect data about physical phenomena in various applications such as habitat monitoring, and ocean monitoring, and surveillance. As an emerging technology brought about rapid advances in modern wireless telecommunication, Internet of Things (IOT) has attracted a lot of attention and is expected to bring benefits to numerous application areas including Industrial WSN systems, and healthcare systems manufacturing.wsn systems are well-suited for longterm industrial environmental data acquisition for IOT representation. The proposed design of Wireless Sensor Network based Life Save System collects the data from the remote sensors and streams the hazardous situation live on a network using Wi-Fi network. Data will be received by through Mobile or any other device connected to Wi-Fi network through Secures Shell (SSH) network protocol. The data enable user to come know about presence of poisonous gases and hazardous situation. The system uses Arduino Uno for logging data, webcam for live video stream Raspberry Pi booted with Raspbian Jessie Lite operating system for Wi-Fi client and live stream. Keywords: Wireless Sensor Network, Wi-Fi, Arduino Uno, Raspberry Pi, SSH, Live Stream 1. Introduction extent. Safety in scavenging and fire fighting areas requires continuous attention of parameters such as temperature and some noxious gases such as CO, CO2, CH4, LPG and Smoke. There is an existence of specific systems or schemes which are put into place for these hazardous environments in order to protect the worker from harm. The higher level term for these systems/schemes is the Occupational Health and Safety. The International Organization of Standardization or ISO have a standard namely the ISO This standard aims to reduce the liability of occupational injuries and diseases not only to benefit the workers but also the economy upon which this work builds [1]. As the situation is hazardous monitoring the data near from the operating location is putting life at risk. The proposed design WSN based Life Save System gives a complete replica of situation in a hazardous place. Without nearing the place, Scavengers and Fire Fighters can know about the hazardous situation by using the proposed system. They can arrange protective equipment, corrective measures accordingly. 2. Effect of Noxic Gases on Human Body A. Effect of Carbon Mono Oxide CO Carbon Monoxide (CO), is often called the Silent Killer because of its ability to take lives quickly and quietly when its victims never even knew they were at risk. It is in detectable to humans, being both tasteless and odorless, and in high enough concentrations it can kill within minutes [2]. Scavengers and Fire Fighters are being exposed to various hazardous situations. In most of the situations they may not know exact scenario of their working place. The protective equipment can protect them up to a certain 487

2 Table 1. Effect of Carbon Monoxide on Human Body Level CO 0 PPM of Health Effects, and Other Information Ambient Air. 9 PPM Safe limit for indoor level 200 PPM Head ache and fatigue PPM 12,800 PPM Nausea, Loss of Consciousness, long time exposure may lead death Death within minutes. B. Effect of Carbon DiOxide CO2 CO2 is a naturally occurring atmospheric gas that is considered safe at levels below 0.5% according to OSHA standards. However, occupational hazards related to CO2 exposure may occur under certain conditions. The American Society of Heating, Refrigerating, and Air Conditioning Engineer, Inc., recommends that indoor air CO2 levels be less than 700 ppm above the outdoor air concentration of CO2 [3]. Table 2. Effect of Carbon Di Oxide on Human Body Level of CO PPM 9 PPM 2,000-5,000 PPM >40,000 PPM Health Effects, and Other Information Normal background concentration in outdoor ambient air Concentrations typical of occupied indoor spaces with good air exchange Headaches, sleepiness and stagnant, stale, stuffy air. Poor concentration, loss of attention, increased heart rate. Exposure may lead to serious oxygen deprivation resulting in permanent brain damage, coma, even death. C. Effect of LPG and Smoke on Human Body LPG in gaseous form acts as a simple asphyxiant and a central nervous system depressant. Over exposure to LPG Can cause lightheadedness and drowsiness. Greater exposure may also cause unconsciousness and death. Contact with the liquid may also cause frostbite and irritation. The permissible limit for LPG is 1000 ppm during 8 hour work shift [4]. The smoke is a mixture of particles and chemicals produced by incomplete burning of carbon-containing materials. All smoke contains carbon monoxide, carbon dioxide and particulate matter (PM or soot). Frequent exposure to smoke for brief periods may also cause longterm health effects. Firefighters, who are exposed frequently to smoke, have been examined for long-term health effects (for example, cancer, lung disease, and cardiovascular disease) of repeated smoke exposures. The findings from these studies are not consistent or conclusive. Some studies show an increased frequency of these diseases among firefighters [5]. 3. System Overview The design uses Arduino Uno for data logging from sensors. The Arduino Uno acts as a sensors node. Arduino Uno sends the data logged from sensors CO (MQ-7), CO2 (MQ-135) and LPG/Smoke (MQ -2) to Raspberry Pi B using serial interface cable. Raspberry Pi B is comes as a low cost handy computer, which runs on variety of RTOS. This Raspberry Pi B makes data logging and remote monitoring much easier through Wi-Fi IEEE Standard. Raspberry Pi B has built-in graphical processor makes video live stream smoother and easier. The video will be recorded using web cam that is fixed on a raspberry pi. The Raspberry Pi B streams the video live on the network through its IP address thus by acting as a webcam sever. The data received by Raspberry Pi B will be saved to a text file the same info will be displayed on a SSH Client device that is connected to a network. This is achieved by Python program that is written on Raspberry Pi B. For accessing the Wi-Fi network Pi is connected to Wi-Fi Adaptor. The proposed design deals with critical parameters which may risk lives and also it uses Wi-Fi network so, the need of network security is inevitable. The proposed design uses Secure Shell Client (SSH) to authenticate user to login over the network. The Raspberry Pi B acts as a SSH Server. Through SSH any device which is on the network and has password can login and get data from the Raspberry Pi B. The live stream video can also be viewed by accessing the IP address of the Pi. 488

3 U=RL/(RL+Rs)*Vc R is an external resistor, Vc is the loop voltage, in other words the power supply voltage of sensor. U is the output voltage of the sensor, Rs is the volume resistivity of the sensor and RL is the load resistance. If the gas concentration rises will lead to decline in Rs. If the value of Rs decreases due to presence of Carbon Mono Oxide the output voltage of the sensor increases. The output voltage varies linearly with the concentration of Carbon Mono Oxide [6]. B. Measurement of CO2 Figure 1. System Overview 4. Design of Data Logging System for Sensors The design uses MQ-7 for Measuring CO, MQ135-for measuring LPG/Smoke, MQ-2 for measuring CO2 and ultrasonic sensor for distance measurement. The sensor modules are connected to Arduino Uno Analog Inputs. Arduino acts as a Sensor Node. Arduino logs this sensor values and communicates serially to Raspberry Pi B+ board. The MQ-135 gas sensor more sensitive to Carbon Di Oxide. The operating voltage of this gas sensor is from 2.5V to 5.0V. The MQ-3 gas sensor has a lower conductivity to clean the air. The atmosphere has polluting gases. The output voltage of gas sensor increases as the concentration of polluting gas increases. MQ-135 gas sensor also can be implemented to detect the other harmful gases such as smoke, benzene. It has different potential to for hazardous gases. The MQ-135 gas sensor is low cost to purchase. A. Measurement of CO Working of MQ-7 Sensor MQ-2-type sensor is a tin dioxide semiconductor gas materials used in the design, surface-ion N-type semiconductor. When at 200 ~ 300 C temperature, tin dioxide adsorption of oxygen in the air, the formation of oxygen anion adsorption so that the decrease in the electron density in the semiconductor and its resistance value increases. When in contact with the smoke, if the potential barrier at the grain boundaries by the modulation of the smoke varies, it will cause changes in the conductivity of the table. [6]. Figure 1.3 MQ 135 Sensor Figure 2. MQ 7 Sensor According to the working principle of MQ-7, the following formula can be obtained Figure 4. Sensitivity Characteristics of MQ 135 Sensor 489

4 The graphic above seems a power function of CO2 So, CO2 ppm=x*(rs/ro)^y By using power regression we can obtain scaling factor (x), and exponent (y), for the Carbon Di Oxide. For calibrating the CO2 sensor, we need to know amount CO2 at the test area. If we know that we can read the resistance output value from the sensor Rs, and we can adjust RL so, as to calibrate the sensor. The known amount of CO2 in the atmosphere is 392ppm at test area. For CO2, the measured points graph and by using power regression we can obtain the function CO2 ppm = (Rs/Ro)^ [7] By using the formulae Arduino Uno was programmed so as to give ppm output of CO2. C. Measurement of LPG/Smoke MQ-2 is sensitive to LPG/Smoke. The MQ-2 is has Sensitive Resistance Rs and Load Resistance R. The MQ- 2 sensor can also measures butane, methane, alcohol & propane. MQ-2 Sensor widely used in industry and domestic for finding gas leakage. The sensitive resistance changes with respect to various gas concentrations. The sensitive resistance may vary from 4.7K, the load resistance is adjustable. The sensitive resistance and load resistance acts like a voltage divider. With output voltage variations we can able to predict the ppm values of LPG/Smoke. For better accuracy if characteristic curve readings for different gas concentrations are found and if curve slope values are to be taken into consideration. In the clean air value of R can be found out during the calibration. The aim of the calibration is to measure R by sampling and by using averaged readings. Once sensor is calibrated, then ppm values of Smoke, LPG can be derived by measuring the output voltage using ADC in Arduino Uno. [8] D. Measurement of Distance Using Ultrasonic Sensor In case of underground operations measurement of depth is necessary and toxic gases may present at any depth. Ultrasonic sensor employed here is to measure depth of the underground. Ultrasonic sensors sends the ultrasonic waves, from the reflected echo waves it calculates the distance using formulae. Distance= (Speed*Time)/2 Speed refers to speed of the air i.e. 340 m/s. For demonstration purpose the design uses 4m range sensor, for real time application the higher range can be used. The sensor sends 40 khz signals for a time period of 10µseconds. 5. Design of network The design uses Raspberry Pi B+, it is a mini computer. The Raspberry Pi booted with Raspbian Jessie Lite Operating System. Raspberry Pi having 512 MB RAM, 4 USB Ports, 40 GPIO pins, Video out pin, 3.5mm Audio jack, SoC Broadcom 2835 and supports Wi-Fi through USB Wi-Fi adaptor, LAN through Ethernet Socket [9]. A. Wi-Fi Network The design uses Mobile Wi-Fi hotspot to set a portable Wi-Fi Network u IEEE standard. The Wi-Fi hotspot is secured with Wi-Fi Protected Access Pre Shared Key (WPA-PSK) client authentication method. The Wi-Fi network Service Set Identified (SSID), Network Security Key is configured with Raspberry Pi B+ module. B. SSH The design uses Secure Socket Shell (SSH) to connect remote computer through a secured access in an insecure network. SSH provides strong authentication and encrypted data communication between devices on the network. SSH is cryptographic network protocol which uses client-server model. The Design uses Raspberry Pi B+ as a SSH-Server. The SSH-Client application is installed in the mobile, which allows user to login to terminal of the Raspberry Pi B+. From the terminal user will able to execute the commands and programs. C. Live Stream Network The video is recorded using Webcam, the webcam communicates to Raspberry Pi B+ using USB Cable [10]. The resolution of the webcam can be fixed up to 6048*4032 and maximum frame rate 30fps. The video is recorded using webcam. The webcam communicates Raspberry Pi B+ using USB Communication protocol. The video is live video is streamed on the network by using Real time Streaming Protocol (RTSP), Real time Transport Control Protocol (RTCP). Open source software package called Motion is installed in Raspberry Pi B+. The motion software is configured to broadcast the live stream through the Raspberry Pi B+ (local host) IP Address on the network and it is configured for video for video codec, frame rate, port and resolution etc. The design uses video codec-mpeg4, port-8081, framesrate-100fps, resolution 640*480. The live stream can be stopped or started from the terminal using SSH- Client on the network. The live video stream can be 490

5 viewed by any device on the network by accessing IP address of Raspberry Pi B+. D. Sensor Data Logging and Monitoring The data from sensors are logged to Arduino Uno. The sensor data is communicated serially to Raspberry Pi B+ using USB serial cable. The Raspberry Pi B+ is configured to run a python program, which get the serial data and displays it on terminal and it writes the same to a specified text file. The text file can be viewed through terminal. Users who are on the same network can login through SSH can view the data by accessing the text file or terminal. The report contains suggestions & sensor data can be accessed by Firefighter or Scavenger by logging into Raspberry Pi B+ terminal from their mobile device. From the report & live stream Scavenger or Firefighters may able to know the hazardous gas levels and hazardous scenario. 7. Results and Observation The video is live streamed on Raspberry Pi B+ IP address. 6. Designed System Figure 6. Screen Capture of Live Video Stream For serial communication interface of Arduino Uno is monitored initially with the Serial Monitor in Arduino IDE (Integral Development Environment). Figure 5. View of Designed System The Raspberry Pi B+ will be connected to Wi-Fi Network that is specified in its configuration. The sensors will be calibrated initially this will be indicated by buzzer. The data from the MQ-7, MQ-135, MQ-2, Ultra Sonic sensors will be logged into Arduino Uno. Arduino Uno is programmed to such as to generate report with respect to the sensor data. The report will be transmitted serially to the Raspberry Pi B+ module. The Raspberry Pi B+ will receive this data. With the python program written on Raspberry Pi B+ it will store the report into specified text file. The Webcam will record the video sends it to Raspberry Pi B+ using USB serial cable. The Pi s USB port receives it. Using configured Motion Software formats the video and streams the live video on the IP address of Raspberry Pi B+ (local host). 491

6 Figure 7. Serial Interface Output Verification The System is kept in a room output received on the terminal of the Raspberry Pi B+ accessed through SSH Client application installed in the mobile. Figure 8. Output received at SSH Client Mobile App Figure 9. Output text file at Room Environment The system is exposed to smoke for 5 minutes, the results obtained in a text file is shown below. The report is stored as a text file in Raspberry Pi B+ memory. The output of the text file, when system is kept in a room. Figure 10. Output text file at Smoky Environment Results are obtained as follows, 492

7 Table 3. Room Environment Parameter Expected Outcome Level Actual Outcome Level CO 0 ppm 0 ppm CO ppm 262 ppm Smoke 0 ppm 0 ppm Table 4. Smoky Environment Parameter Expected Outcome Level Actual Outcome Level CO 2-3 ppm 2 ppm CO ppm 372 ppm Smoke 5-7 ppm 5.83 ppm From the above results the system proves 80%, 71%, 95% (approx.) accuracy to gases CO, CO2 & Smoke. The system over all accuracy 82% (approx.) 8. Future Aspects The essential information for firefighting is fire source, with the help of thermal imaging it is easy to understand about fire source, fire spreading direction. An additional thermal imaging system is going to be used in future. The cloud computing makes monitoring easy, also it is easy to integrate multiple sensor nodes. The system can also give accurate results by using data analytics and cloud computing techniques. In future system incorporating cloud computing & data analytics can be implemented. 9. Conclusion Therefore WSN based Life Save System was designed. The system is compact and can be fixed to any drone or mobile robot to monitor CO, CO2 gases and LPG, Smoke levels. The system will communicate data from sensor node and stream live video to user s mobile through Wi- Fi. The system will give exact scenario through the sensor data and video live stream. The can also be implemented in Mining and Chemical Industries for monitoring hazardous zones. References [1] Valdo Henriques, Reza Malekian Mine Safety System Using Wireless Sensor Network IEEE Access vol [2] Carbon Mono Oxide Health Issues retrieved from CarbonMonoOxide.com [3] U.S Department of the Interior Buraue of Land Management NEPA (National Environment Policy Act) [4] U.S Department of Health and Human Services- National Institute of Occupational Safety and Health Occupational Health Guidance for LPG-Retrieved from on 10th March [5] New York State Govt. Department of Health- Exposure to Smoke from Fires retrieved from the source on 10 th March oke_from_fire [6] Pan Jiru College Students Shenyang University of Technology Shenyang,China Design of Intelligent Monitoring System Third International Conference on Instrumentation, Measurement, Communication and Control [7] Davide Gironi - Design of cheap CO2 Monitor Using MQ135 Sensor Using AVR ATMega [8] Sandbox Electronics MQ-2 Smoke/LPG/CO Gas Sensor Module Product Datasheet [9] Kristoffer O. Flores, lsidro M. Butaslac, Jon Enric M. Gonzales, Samuel Matthew G. Dumlao, Rosula S.J. Reyes Precision Agriculture Monitoring System using Wireless Sensor Network and Raspberry Pi Local Server Region 10 Conference (TENCON), IEEE, Nov [10] Neel Oza, N. B. Gohil Implementation of Cloud Based Live Streaming for Surveillance 2016 International Conference on Communication and Signal Processing (ICCSP) April [11] Xing-qian LI,Yue ZHANG, Hong-wei ZHAO, XinDING Zhi-ping SUN Iot Family Robot Based on Raspberry Pi International Conference on Information System and Artificial Intelligence (ISAI) June 2016 [12] Olimex Ltd,MQ-2 Semiconductor Sensor for Combustible Gas- Data Sheet [13] Olimex Ltd,MQ-135 Gas Sensor Technical Data Sheet [14] TP Link WN821N Wi-Fi Adaptor Specification retrieved from 493

8 WN821N(US)_V5_User_Guide_ o.pdf on 08 March [15] Yogendra Singh Dohare, Tanmoy Maity, P. S. Paul, Hanuman Prasad Smart Low Power Wireless Sensor Network forunderground Mine Environment Monitoring International Conference on Recent Advances in Information Technology (RAIT) March [16] Rajesh, M., and J. M. Gnanasekar. "Path observation-based physical routing protocol for wireless ad hoc networks." International Journal of Wireless and Mobile Computing 11.3 (2016): [17] S.V.Manikanthan and T.Padmapriya Recent Trends In M2m Communications In 4g Networks And Evolution Towards 5g, International Journal of Pure and Applied Mathematics, ISSN NO: , Vol-115, Issue -8, Sep [18] T.Padmapriya and V.Saminadan, Utility based Vertical Handoff Decision Model for LTE-A networks, International Journal of Computer Science and Information Security, ISSN , vol.14, no.11, November

9 495

10 496