Wireless Sensor Networks for Environmental Monitoring Applications
Chapter One
Objectives of the Study
GOAL
The aim of this project is to set up a robust WSN for Environmental Monitoring Application of ecological data (biodiversity).
SPECIFIC OBJECTIVES
- Set up a cloud storage service for collection of real-time biodiversity data (temperature and humidity) and synchronize the data between the BS and cloud
- Designand implement a communication protocol for collection of tracking
- To explore the techniques of data collection and aggregation based on similarity and redundancy of
- To implement a remote web application service for monitoring of real-time sensor node failure.
CHAPTER TWO
CONCEPTUAL ANALYSIS
WIRELESS SENSOR NETWORK
WSN application has grown widely in the last decade. Many industries, researchers and engineers work on this novel technology by applying it in civilian and military sectors that use many nodes. Many applications have been developed and deployed for the purpose of solving numerous problems in data acquisition and environment control. These node networks must have the capacity, via universal wireless, of sensing, processing and communicating physical parameters such as pressure and temperature [3].
A WSN is described as networked nodes that jointly sense and manipulate data, enabling interaction between humans or computers and the immediate environment (Bröring in [4]). WSNs today include sensor nodes, actuator nodes, gateways (Internet connectivity) and clients. It also entails a large number of sensor nodes randomly deployed indoors or near the monitoring area (sensor field), which form networks through self-organization. Operationally, sensor nodes observe the data collected and transmit to other sensor nodes by hopping. For the period of transmission process, sensor data may be moved by multiple nodes to get to a gateway node after multi-hop routing, and eventually end with the management node through the Internet or satellite as the case may be. The user configures and manages the WSN with the management node, broadcasts monitoring jobs and collects the monitored data [5].
WSNs are critically resource inhibited by limited power supply, memory, performance and communication bandwidth [6]. While initial sensor nodes were resource-constrained with limited capabilities, recent improvements in sensor hardware technology have made it possible to have better processing power sensors, memory and protracted battery life [7]. But Mohammad in [8] cites Heinzelman and states that the energy consumption of radio communication largely depends on the number of bits of data that can be transmitted within the network.
Fischione in [9] states that WSNs make the Internet of Things (IoT) possible and defines it as a networked wireless system for computing, transferring and receiving nodes meant for interaction, controlling, sense and activation.
He further stated the following as characteristics of WSNs:
- Battery-operated nodes;
- Short range wireless communication;
- Mobility of nodes; and
- No/limited central
TYPES OF WSN
Ado et al. in [10] list diverse types of WSN as follows:
TERRESTRIAL
This is largely used in the field of environmental monitoring and poses a challenge to the sustainability of the network in terms of energy management [11].
UNDERGROUND
Sensors are buried underground using wireless technology to enable them to communicate. It is used for agricultural purposes to monitor conditions in the soil. It has a land node to transfer sensed information from the underground nodes to the BS [12]- [13].
UNDERWATER
Ado et al. cited by Potdar et al. [10] explain that this type of WSN still imposes serious research challenges due to the hostility of the deployed environment to the nodes which are usually meant for exploration. Here there is recurrent loss of signal, propagation delays and synchronization problems are high.
MULTIMEDIA
This monitors real-time data like images, videos and audio. The sensors use camera and microphones. The problem here is high energy consumption according to Misra in Ado et al. [10].
MOBILE
The mobile nodes have the ability to autonomously reorganize the network and communicate with the physical environment. This network is more flexible than the static sensor networks because it has the ability to improve coverage, energy efficiency, superior channel capacity, and so on [14].
CHAPTER THREE
METHODOLOGY (ANALYSIS AND SOLUTION DESIGN)
These topics are discussed: design of the study, area of the study, instrumentation for data collection and method of data transmission.
The challenging factors in WSN for Environmental Monitoring Application using Raspberry Pi, Arduino and XBee DTH11 temperature-humidity sensors as discussed in the previous chapter, have posed a lot of difficulties especially as regards online real-time node failure detection, user- friendly data presentation and inability of Raspberry Pi (BS) to store large data over a long period of time due to small memory capacity. The storage capability of Raspberry Pi in handling accumulated data was extensively discussed and cloud storage facility was proposed by Sheikh and Xinrong in [33] as a long-standing solution.
SPECIFICATIONS
The pragmatic aim of this experimental research was to implement a communication protocol for data collection, set up a cloud storage service to sync data online in real time from a remote BS and constantly monitor real-time node failure through a web interface.
SCOPE / AREA OF STUDY
Cloud storage service as an alternative solution for storing and managing of large data that may not be contained in Raspberry Pi was proposed in [33] by Sheikh and Xinrong. In their paper, they suggested integrating cloud service as a long-standing solution to maximize small storage capacity of Raspberry Pi. This proposal was implemented in this project to solve the long-standing challenge. While there may be other challenges in environmental monitoring, this research limits its scope to monitoring and collection of biodiversity data like temperature and humidity on terrestrial (above ground) [58] and management of the data over Amazon Cloud. The Amazon Cloud service (AWS) was used because the platform offers a free service.
CHAPTER FOUR
IMPLEMENTATION
The implementation of the concepts and methodologies detailed in the last chapter comprised over 1000 lines of C, Python, PHP, JavaScript and HTML code. The difficulties encountered in WSN for environmental application in the past, that led to this research as regards online real-time node failure detection, user-friendly data representation and using cloud as a secondary storage server to store large accumulated data that may exceed Raspberry Pi (BS) storage capacity have been solved in this project. There was a clear-cut defined communication protocol to transmit data (TX) and Receive data (RX) encoded in the AT Mode function set.
AWS EC2 cloud service was chosen because it was free to use and offers a reasonable storage capacity with other flexibilities. This chapter contains the full description of the implementation of the main functions and data structures that sums up the whole system.
CHAPTER FIVE
TESTING
WIRELESS SENSOR DATA COLLECTION
The WSNd data gathering implemented with Arduino code is one of the key breakthroughs in this project because it gave a basis for the work to be expanded to a further stage. The program was tested and the following data output was received from an air-conditioned room. This is a controlled environment but this node can work in any terrestrial habitation. The sensor node is strong enough to withstand any harsh weather. The data below was collected on 23 May 2016 in an air-conditioned office.
CHAPTER SIX
CONCLUSION
We have designed and implemented a WSN for Environmental Monitoring Application (WSNEMA) that was leveraged on cloud as its main storage system for a real-time online ecological data management/storage system. Also a tool for detecting real-time Online Wireless Sensor Node failure was implemented.
KNOWN LIMITATION
The present implementation of WSNEMA only supports the collection of temperature and humidity data. It is important to expand the system and modify it to collect another type of environmental data like soil PH, wind (speed and direction), pressure etc.
FUTURE WORK
Further research should focus on developing a generic platform that will be simple for anybody to use. In most cases those who handle environmental data are not programmers and the existing platform requires programmers for implementation. Generic platforms with all sensor libraries installed should be able to allow anybody to buy any terrestrial sensor and select the type on an API (GUI) to configure and the needed code for the running of the sensor(s) automatically.
Due to data loss during packets transmission that leads to data corruption, there is need to further enhance this work and implement an API MODE for both the XBee router and the XBee coordinator to make data reception and transmission very accurate. (i.e. by defining Source Address, Destination Address, Message Part, Check Sum, etc.).
SUMMARY
In the introduction we highlighted some of the problems facing WSNEMA, which include but are not limited to first, lack of immediate detection of sensor node failure on an online real-time basis; second, the issue of small storage space in Raspberry Pi makes it difficult to store large
volumes of data accumulated during data gathering; and third, implementing a user-friendly way of data presentation. We have through this project solved the issues being experienced in the current state of the art of the WSEMA.
This work focused on how to solve the challenges mentioned in Chapter 1 of this project, relating to collection of environmental data (temperature and humidity) using Arduino UNO, Temperature-Humidity Sensor, Raspberry Pi and ZigBee (XBee S2). It could also serve as a prototype for collection of other environmental data that requires wireless sensor node communication. We implemented a scalable structure that can accommodate more sensor nodes where expansion becomes very necessary most especially when considering data aggregation.
Although substantial work has been done in designing and implantation of WSNEMA, there are yet unanswered questions. Taking the research further will not only advance the field of WSN but will complement efforts in understanding and making decisions in combating danger inherent to climate change.
REFERENCES
- Arnold O. Allen (1994): ‘‘Computer Performance Analysis with Mathematics’’. Academic Press, Roseville California,. pp 1. ISBN 978-0-12-051070-2.
- Agarwal, A.; Levy, M.,(2007): ‘The KILL Rule for Multicore’, Design Automation Conference, DAC ’07. 44th ACM/IEEE, vol., no.4, pp.750-753, 4-8 June 2007
- Chapman, G. Justand R. Van de Pass,(2008): ‘using OpenMP, Portable Shared Memory Parallel Programming’, Massachusetts Institute of Technology, pp.4, 16-35
- Conallen, Jim (2000).‘‘Building Web Applications with UML’’. Addison Wesley. p.147. ISBN 0201615770.
- Carpenter R. E.,(2007): “Comparing Multi-Core Processors for Server
- Virtualization”, Intel Corporation.http://www.multicoreinfo.com/research/papers/whitepapers/multicore_virtualization.pdf
- Domeika M., (2009): “Evaluating the Performance of Multi-Core Processors”,http://www.embedded.com/design/embedded/4008794/Evaluating-theperformance- of-multi-core-processors-Part–1
- Faxen K.,Benggttsson C.,Bronsson M., Grahn H.,Hagersten E., Jonsson B., Kessler C.,LisperB., Stenstrom P., Svensson B.,(2008)”Multicore Computing-The State of Art”, p.3. URL: http://eprints.sics.se/3546/
- Jacobsen, Ivar; Magnus Christerson, Patrik Jonsson and Gunnar Overgaard (1992). ‘‘Object Oriented Software Engineering’’. Addison-Wesley ACM Press. pp. 77–79. ISBN 0-201-54435-0.
