Project Topics
Get Complete Project Material File(s) Now! »
Internet of Things
The Internet of Things (IoT) refers to uniquely identifiable objects and their virtual representations in an Internet-like structure. IoT is getting great interests in recent years in the scenario of modern wireless telecommunications. The concept of IoT can be considered as the result of convergence of different three visions [7]. From “Things oriented” perspective, IoT emphasizes the usage of Radio-Frequency Identification (RFID) tags in things, so as to make everything identifiable. From the “Internet-oriented” perspective, supporters try to incorporating IEEE 802.15.4 into IP architecture. From the “Semantic -oriented” pers pective, it is said that things can be reasoned by the Web. Recently, many researchers focus on combining the passive RFID tags with sensing technologies. Several solutions have been proposed, one of which is the WISP (Wireless Identification and Sensing Platform) project. The WISP project is conducted by the Intel Labs, aiming to develop wireless identification and sensing platforms (WISP). WISPs harvest power from the reader’s querying signal of standard RFID readers. WISPs have been well used in measuring quantities in certain environment.
Internet Ø [8] reduces the complexity of the IP stack to achieve a protocol designed to route ‘‘IP over anything”. Recent years , a lot of attention has been paid to the integration of RFID tags into IPv6 networks [9] and methods that integrate RFID identifiers and IPv6 addresses have been proposed.
The consortium CASAGRAS (Coordination and Support Action for Global RFID-related Activities and Standardization) [10] proposed an IoT vision statement that is far beyond the “RFID centric”. Its members emphasize in ‘‘a world where things can automatically communicate to computers and each other providing services to the benefit of the human kind”.
In IPSO [11], the authors talked about incorporating IEEE 802.15.4 into the IP architecture and Internet Protocol for smart objects. In accordance with the IPSO Alliance vision, making IoT a reality with IP is possible. It seems that the full deployment of the IoT paradigm will be automatically enabled, through a wise IP adaptation and by incorporating IEEE 802.15.4 into the IP architecture, in the view of 6LoWPAN (IPv6 over Low power Wireless Personal Area Networks).
A further vision correlated with the IoT is the so called ‘‘Web of Things” [12], according to which Web standards are re-used to connect and integrate into the Web everyday-life objects that contain an embedded device or computer.
Semantic Web
The World Wide Web (WWW) develops in an amazing speed ever since it was born in 1989, bringing about the information explosion. However, the majority of information on the Internet is presented for humans to read, computers cannot catch the meaning of the information in web pages. So as to provide a machine understandable representation of web information, Tim Berners-Lee, the inventor of the World Wide Web, puts forward the idea of the Semantic Web [13]. He said « the Semantic Web is an extension of the current web in which information is given well defined meaning, better enabling computers and people to work in cooperation ». The Semantic Web brings structure to the meaningful content of Web pages, so that software applications that roam from page to page can readily carry out sophisticated tasks for users [13].
To develop Semantic Web, Berners-Lee proposed the Semantic Web architecture [13], which shows the technologies, tools, and standards used and their semantic levels in the Semantic Web. Figure 1-10 illustrates this architecture.
The Semantic Web architecture starts from Unicode and Universal Resource Identifier (URI). Unicode defined a unique number for each character, which do not relay on the underlying platform, program, or language. A URI is a formatted character string playing the role of identifying abstract or physical resource [14].
The Extensible Markup Language (XML) is a markup language that defines a minimum number of rules for the representation of information by defining own tags of users and add arbitrary structure to the document. XML was designed to carry data in a software- independent and hardware-independent way, meanwhile human- readable and machine-readable [15].An XML schema, which is written in XML, is used to describe a type of XML document, defining the legal building blocks, expressing the structure and content of that XML documents. XML Schemas support for data types, as one of the greatest characteristic, which make it easier to describe permissible document content, such as the elements, attributes, and data types [16].
Sensor Web Enablement
A sensor web refers to the sensor networks that can be connected to the Web. And using standard protocol, the sensor data can be accessed by application program interfaces (APIs). Sensor Web Enablement (SWE) builds a unique framework of open standards for operating Web-connected sensor systems of all types [26]. SWE standards are developed and maintained by members of Open [39]. The functionality of a sensor Geospatial Consortium (OGC), which is an international industry consortium focusing on developing open standards for geospatial and location services [27].
Deciding the capabilities and measurement quality of a sensor. Accessing to sensor parameters.
Acquisition of the real-time or time-series observations.
Publishing of and subscription to alerts to be issued by sensors or sensor services.
Several standards defined for sensor web are put forward to enable the sensor web through the SWE efforts. Standards make it possible not only to develop the applications in an efficient way, but also to simplify the maintenance work. What is more, the standards also make it easy to do future expansions. The main OGC Standards that have been built and adopted in the SWE framework include the following:
Observations & Measurements Schema (O&M) – providing a standard model used to represent and exchange results that have been observed from sensor system and sensor-related processing, and structure those results in XML.
Sensor Model Language (SensorML) – providing an information model used to discover and task sensors, and exploit the observations of sensors. SensorML defines everything as process models, which defines the inputs, outputs, parameters, and method for that process. With the help of XML Schema, SensorML can provide additional metadata, be helpful for nodes discovery, system constraints identifying, properties describing.
Transducer Markup Language (TransducerML or TML) – providing a method for describing information about hardware response characteristics of transducers. TML also provides an efficient and effective way to capture, transport and archive transducer data, using the XML descriptions. Sensor Observations Service (SOS) – providing an open interface for a web service to get observations of a sensor system. SOS makes it possible to manage deployed sensors and to retrieve sensor observation data.
Sensor Planning Service (SPS) – providing an open interface for requesting, determining the feasibility of collecting data from sensors and submitting the collection requests.
Sensor Alert Service (SAS) – allowing sensors nodes to publish and subscribe alerts from other sensors or advertise its describing metadata respectively.
Web Notification Services (WNS) – Standard web service interface for conducting asynchronous messages or alerts delivery with one or more other services SAS and SPS web services and other elements of service.
An exemplary case-study of the application of SWE in a real-world hydrological deployment scenario is shown in Figure 1-16 and is based on what has been described in the paper “Sensor Web in Practice ” [28]. The SWE services are applied to manage a network of hydrological sensors (e.g., water gauges, weather stations, or cameras observing critical facilities) by providing access to sensor data, by realizing event handling, and by enabling interoperable tasking of sensors. The described deployment can be easily adapted to other real-world deployments of sensor networks.
Main content and organization of the thesis
This thesis talks about the method to solve the problems in WSNs. We introduce how to design and implement a data publishing system of the WSNs.
Then we talk about how to integrate Semantic Web technology to WSNs, at last we introduce a method to publish data of WSNs without sink node.
The chapter 1 of the thesis introduces the background and related researches. In chapter 2, we talk about the requirements of the system in the order of the problems we want to solve. In chapter 3, we explain how to publish data of WSN in detail, and introduce how to integrate the Semantic Web technology. Besides we analysis the “sink neighborhood problem” and introd uce our method of publish data without using sink node. In chapter 4, we talk about the implementation and results of the tests applied to the proposed system.
The goal of the system
A Wireless Sensor Network (WSN) is a network which consists of a large number of wireless sensor nodes, which are small autonomous devices that have wireless communication ability to automatically organize themselves to form a multi-hop network. WSNs provide a promising infrastructure to sense their surrounding environmental phenomena through the set of transducers imbedded or plugged on the sensor board [41].
Since WSN is quite different from the Internet, publishing data of WSN to Internet is the first problem when doing WSN development. A WSN may apply a great amount of wireless sensor node for different end users, so the convenience that a user finds his interesting data from so many nodes and sensors is an important factor to be taken into consideration. To fulfill the demand, we will develop a data publishing system of WSN, which makes it possible for users to check information of the whole network and each wireless sensor node. What is more, we enhance the system with subscribing function, which enable user to subscribe his interested sensors, a little bit like the working process of Twitter, so as to convenient for checking data.
The development of WSN requires the abilities that enable the interoperability of sensor data to support re-use of existing sensor networks, and relating the sensor data with stored data. On the other hand, the Semantic Web brings the data and documents with machine-readable descriptions and makes links between different data source on the Web which enables the interoperability. So in order to fulfill this demand, we introduce Semantic Web technology into WSN as the second goal of our project.
Traditional WSNs use sink nodes to interact with remote users, to collect data and assign tasks. The sink nodes are a kind of special sensor nodes, which are considered as rich in resource of energy and memory, and more powerful processing ability. These sink nodes may also have special hardware components to support communication with the Internet. Because of wireless sensor nodes are always deployed in large numbers and in inaccessible environments, to recharge or replace their batteries is generally impossible. Whenever a node uses up its energy, it is unable to fulfill its detection and communication responsibilities any longer, so this node is considered as a “dead” node [42]. Since collected data from all the sensors are delivered to the sink, nodes that are closer to the sink have to bear greater burden. More specifically, when a sink is positioned statically, the neighbor nodes of the sink, which communicate directly with it, are likely to exhaust their energy more quickly than the other nodes. Not only because they spend energy on propagating data to the sink node of their own, but also for relaying data from other nodes to the sink. This problem was termed the “sin k neighborhood problem” in [42]. When all the neighbor nodes are dead, the sink will get isolated from the rest of the network, and the WSN will be inaccessible, though most of the wireless sensor nodes are still at work. Besides, an ordinary node has to relay on sink node to publish their data, this would reduce flexibility on using the sensing node. So, try to find a solution to solve the “sink neighborhood pro blem”, to be further, try to find a way publishing data of WSN without using sink node, is the final goal of the project.
To make a summary, there are three goals need to be realized in this system, including:
1) Develop a data-publishing system of WSN.
2) Find a way to publish data of WSN based on the Semantic Web technology.
3) Researching on how to publish data of WSN without using sink node.
Requirement of WSN data publishing system
Since the project has three goals, so we discuss the requirements of each goal separately. In this part, we mainly talk about the requirement of the data publishing system of WSN.
Dataflow diagram
To analysis the requirement, we first draw out the dataflow diagram at the primary level to get a basic knowledge about the flow-in and flow-out data of the system.
Figure 2-1 shows the data flow diagram of this data-publishing system based on the goal of the system. We know that data is collected by wireless sensors and transmitted to the system. After some process, the data, at one hand, is sent to a database and stored into it. At the other hand, the data is published to the users. When users make use of the system, they can either view the data collected by the wireless sensors, or send commands to the system. According to the commands, the system will decide whether to send the commands to the database to ask for data, or send to the wireless sensors to set the parameter.
The functional requirements
To make a reflection on the system, the desired WSN data-publishing system is a Browser-Server architecture Web application up on the WSN. The functionality of server is to gather and store data from the distributed wireless sensors, and provide users the service to access those data, while the Browsers are used to observe and request information from server.
At the server point of view, how to obtain the data from those wireless sensors and store those data detected by those sensors is the key part of the system. What is more, as a WSN, the characteristics to self- organizing multi-hop network should be enhanced by the up layer data-publishing system, as well as the energy and task managements.
Since different users may get access to the system with different level of requirements, a user management sub-system should be developed as a main component of the system, to distinguish different users, so that to provide them suitable permission to the system. It is not hard to imagine that there should be three kinds of users, namely, the administrator, the registered user, and the visitor.
To fulfill the data-publishing system, the most important part of the system is how to display the collected data in user-friendly pages to different kinds of users. Because administrators are more concerned about the status of the WSN, and how to manage the system, and set the parameters of each wireless sensor node, the functions available for administrators will be the superset of the functions for registered users. These latter are more interested in the data collected by the wireless sensors, especially the sensors that they have subscribed. Let alone the visitor who have fewest functions demands.
To make a summary, the WSN data-publishing system that needs to be realized should have the functions, including: sensor detection, sensor management, data collection and storage, data publishing, user management, data subscribing, and error reporting.
Sensor detection function
Since a WSN is a self-organizing node network, when the management system just starts to run, it should detect the available wireless sensor nodes and their location in the detected region. Since a wireless sensor may move itself away from its initial location, or it may run out of energy, or because the environmental condition of the region of deployment is very poor or complex, a sensor can be “dead” and out of connection for various reasons. L ikewise, new sensors could be deployed progressively in the same area. So, the dynamic changes require the system be sensitive about the sensors’ leaving and joining in actions. Meanwhile, the data-publishing system should be aware of the status of a sensor.
Sensor management function
The data-publishing system should have the ability to control the behavior of sensor node. When a new wireless sensor node joins the WSN , or when an in-used sensor node changes its sensing task, the data-publishing system could set new parameters for the sensor node, so as to control what kind of phenomenon to sense and how often the sensor node do the sensing work. What is more, the data-publishing system could have the ability to control the routing strategy, the energy conservation strategy, and so on, which goes beyond the requirements of this project. In practice, this sensor management work is carried out by the administrators of the system.
Data collection and storage function
After the WSN is deployed, wireless sensor nodes may begin to sense its surrounding environment, and send the collected data to the data-publishing system. A wireless sensor node may have different transducers and all it sensed data could be collected by the data-publishing system, regardless of what kind of data have been sensed. After the data is received, the system will transfer those data into a MySQL database management system (DBMS) to store them.
In this system, we assume that all the wireless sensor nodes could sense temperature and its battery volume.
User management function
WSNs are developed for human beings to collect information about some phenomena in a monitoring region. Users of the data-publishing system would be sorted into three kinds. They are administrator, registered user and visitor. Administrators have the authority to send commands to control the WSNs, to add and delete registered users. Visitors can just look at the data of a public sensor node, if they want to subscribe some certain wireless sensor node’s data, they must register to become a registered user.
Table of contents :
CHAPTER 1 INTRODUCTION
1.1 BACKGROUND
1.2 THE PURPOSE OF PROJECT
1.3 THE STATUS OF RELATED RESEARCH
1.3.1 Wireless sensor network
1.3.2 Internet of Things
1.3.3 Semantic Web
1.3.4 Linked data
1.3.5 Sensor Web Enablement
1.3.6 Semantic Sensor Network
1.4 MAIN CONTENT AND ORGANIZATION OF THE THESIS
CHAPTER 2 SYSTEM REQUIREMENT ANALYSIS
2.1 THE GOAL OF THE SYSTEM
2.2 REQUIREMENT OF WSN DATA PUBLISHING SYSTEM
2.2.1 Dataflow diagram
2.2.2 The functional requirements
2.2.3 System structure chart
2.2.4 Use case diagram
2.2.5 The non-functional requirements
2.3 REQUIREMENT OF SEMANTIC BASED WSN
2.4 REQUIREMENT OF SINK EXEMPTED WSN
2.5 BRIEF SUMMARY
CHAPTER 3 SYSTEM DESIGN
3.1 DESIGN OF WSN DATA-PUBLISHING SYSTEM
3.1.1 ZigBee network management
3.1.2 ZigBee network data storage
3.1.3 Publishing and subscribing function Design
3.1.4 Database design
3.2 DESIGN OF SINK EXEMPTED WSN
3.2.1 Related work
3.2.2 Challenges to exempt sink nodes
3.2.3 Architecture of sink exempted WSN
3.2.4 Hardware design
3.2.5 Protocol stack
3.2.6 Data transmitting process
3.2.7 Data receiving process
3.2.8 Problems remained
3.3 KEY TECHNIQUES
3.4 BRIEF SUMMARY
CHAPTER 4 SYSTEM IMPLEMENTATION AND TESTING
4.1 THE ENVIRONMENT OF SYSTEM IMPLEMENTATION
4.1.1 Hardware environment
4.1.2 Software environment
4.2 KEY PROGRAM FLOW CHARTS
4.2.1 Flowchart of data publishing process
4.2.2 Flowchart of user subscribing sensor data
4.3 KEY INTERFACES OF THE SOFTWARE SYSTEM
4.3.1 The mainpage of the system
4.3.2 The sensor page
4.3.3 The personal page of a registered user
4.3.4 The administrator Page
4.4 SYSTEM TESTING
4.4.1 Tests on the mainpage
4.4.2 Tests on the sensor page
4.4.3 Tests on the personal page of a registered user
4.4.4 Tests on the administrator page
4.5 BRIEF SUMMARY
CONCLUSION
REFERENCES
