Home

Wireless Sensor Networking with Lab

1. Figure 2 1 Detail structure of WSN 5 Classical Infrastructure is the existing network structure which may or may not be connected to the internet Simple private network of organizations or industries used for local communication and data exchange can be considered as classical structure Sensor networks as defined earlier are spatially distributed autonomous sensors which are connected to each other forming a network topology ad hoc Star Mesh etc The groups of sensors are connected to AP which ultimately connect sensors to end users via some sort of display units or computers End user can monitor and control the process if required There are three APs interconnected to each other in Figure 2 1 Each AP is responsible for traffic and management flow for their group of sensors One AP is connected to end user via classical infrastructure Such that information from all nodes can be accessed by end users 12 Number of sensor nodes and AP differs as per user s interest of measurement environment and coverage area Deploying mesh topology one should be very careful to know the supportability of multi hop structure node talking to adjacent node by selected devices For example NI WSN 3202 and Wireless HART support internode communication whereas nodes like NI WLS 9163 can be configured only in star topology Advantages of multi hop network topology are it helps to increase the distance of communication and adds redundancy feature
2. It is also true that use of multiple AP can also increases distance in star network topology 2 1 Standard Transmission Protocol The scalar data such as temperature humidity pressure vibration have low sampling rate Vibration data have relatively high sampling rate compared to others However they are considered to be less as compared to video and picture data WSNs mainly deal with such data having less scan rate sampling rate The advantage of low scan rate data is that WSN can be designed based on low power and low bandwidth transmission protocols It gives clear concept of why ISM band is deployed in WSN as transmission protocol ISM band is license free 2 4 GHz band used by Wi Fi Zig bee Bluetooth and Wireless HART 2 1 1 Bluetooth Bluetooth is a low powered low cost and short range protocol governed by IEEE 802 15 1 standard Bluetooth is promoted by Bluetooth Special Interest Group SIG 6 There are 5 different versions of Bluetooth namely V 1 0 b V1 1 V1 2 V2 0 and V 3 0 First three versions use Gaussian Frequency Shift Keying GFSK as a modulation scheme and fourth version uses GFSK a 4 Differential Quadrature Phase Shift Keying DQPSK and 8 Differential Phase Shift Keying 8 DPSK as modulation scheme The maximum data rate supported by V 1 0 b V1 1 V1 2 is I Mbps V2 0 supports up to 3 Mbps and latest version V3 0 supports up to 24 Mbps Maximum transmission range of Bluetooth is 100m Technical overview of Bl
3. WSN consists of at least two nodes communicating each other through RF medium However the numbers of nodes can be more than two which depends upon the measurement strategies and environment WSN has attracted the attention of both the academic and industrial research companies from late 90 s Some famous smart sensors prototype like Berkeley motes and Smart Dust has been designed and implemented in such areas 1 Smart sensors nodes are also commercialized by Crossbow Philips Siemens and National Instruments Since there are lots of companies involved in designing and developing sensor nodes a considerable amount of standardization is required in WSN field For example National Instrument NI WLS 9163 nodes are standardized with Institute of Electrical and Electronics Engineering IEEE 802 11 protocols 2 and NI WSN 3202 is standardized with IEEE 802 15 4 protocols 3 as a transmission medium WSN has been applied successfully in industries health and environment sectors It possesses some merits and demerits Two of the major disadvantages are security and coverage area or transmission distance Since RF medium is used the attackers can easily attack the physical medium and can extract valuable and sensitive information Secured network is the need for today s applications These days many security measures are practiced so as to provide a hacker free network There are lots of implementation issues like deployment m
4. 2005 37 Khodadoustan S Hamidzadeh M Tree of Wheels A New Hierarichal and Scable Topology for Underwater Sensor Networks Sharif University of Technology Iran 2011 38 Dong B A survey of Underwater Wireless Sensor Networks localization system design Texas A amp M University USA 2007 39 National Instruments Why Am I Losing Data When My Node Indicates an Excellent WSN Link Quality May2010 http digital ni com public nsf allkb COBS57DBB7A7512C0862575EF00538C6C09 Mar 2011 40 Romer K Mattern F The Desing Space of Wireless Sensor Networks Institute for Pervasive Computing ETH Switzerland 2004 41 Molish A F Wireless Communications John Wiley amp Sons Ltd First Edition England 2005 386 387 42 Chen M Kwon T Yuan Y Leung V C M Mobile Agent Based Wireless Sensor Networks JOURNAL OF COMPUTERS Apr 2006 1 14 21 43 Kalita H K Kar A Wireless Sensor Network Security Analysis Internations Journal Of Next Generation Networks Dec 2009 1 1 10 44 Bakken T Pant R B Xie P Shrestha A Wireless Sensor Networks Using NI Modules Telemark University Collge 2010 Norway 45 Mohanty P Panigrahi S Sarma N Satapathy S S Security Issues in Wireless Sensor Networks Data Gathereing Protocols A Survey Journal of Theoritical and Applied Information Technology 2010 14 26 46 Kaplantzis S Security Models for Wireless Sensor Networks Monash University Australia Mar 2006 47 Wood A D Stankov
5. Life Time In WSN the sensor nodes are generally placed out in the field and unattended for months or years So WSN deployed are expected to have certain life time Energy supplied to the nodes 92 can either be managed locally by using battery or managed by supplying external power supply WSN in a real time system and security system demands every node must last for many years Single node failure may lead vulnerability in the networks Apart from above mentioned implementation issues factors like sampling rate also play vital role in WSN implementations 7 9 Sampling Rate All sensor nodes have their own sampling rate specification Sampling rate 1s the time used by sensor to measure the physical quantities Generally sampling rates are less if a single node is considered but 1t has considerable effects when multiple nodes are implemented In WSN parent node must handle data from its child node If there are 40 child nodes then parent node is responsible to transmit and receive 40 times as much as it would have been a single child node Due to this sampling rate has more impact on topology design For example National Instrument device NI WSN 3202 has the sampling rate of I samples sec 3 If 20 such nodes are connected to gateway then overall sampling rate will be 20 samples sec which may not be practically feasible Another problem would be if high sampling rate data like picture video and vibration are to be measure
6. WSN can be combined with Global System for Mobile GSM or internet for telemedicine purposes Structural monitoring application WSN is used to detect the damage in buildings stadium tower bridge ships aircrafts and vehicles This is done by monitoring the response to ambient or force excitation created on the structure 25 f Military application Deploying WSN in battlefield provides the awareness of situation in the field This helps operational forces to make valuable decision Aircraft is used to scatter micro sensor nodes randomly in the battle field and intelligence information are received in base camp 4 2 Advantages WSN has wide range of operational environment It provides advantages in cost size power flexibility and intelligence compared to wired sensor networks Some advantages of WSN are described below a b c d e It avoids lots of wiring and risk of cutting the bus connecting sensors that persists in field bus or other wired technologies WSN uses Micro Electro Mechanical Systems Sensors MEMS MEMS technology has low cost small size and low power requirements 21 In situations where cabling is very difficult dangerous and impossible WSN can be used to avoid cable installations Relocation and restructuring of wireless network is easier than that of wired network Radio signals can easily penetrate the buildings and walls It avoids the need of drilling and ma
7. around 23 C occurred least of the time of observations The maximum temperature for sensor node I was measured to be 23 3 C at 11 sample time and minimum temperature was measured to be 20 9 C at 836 sample time Mean value of temperature was found to be 21 77 C Histogram for sensor node 1 suggested that the maximum occurrence was 21 C and the minimum occurrence was 22 C There were much variations of temperature in sensor node 2 than sensor node 1 Statistical scopes for both sensors consist of green red and white color These represent minimum value maximum value and arithmetic mean value respectively 6 2 Phase lIl Multi hop Network Topology Sensor node with external power supply was placed stationary at 23 1 meters away from gateway 23 1 meters distance was found in first phase of lab work and that was the maximum distance supported by one sensor node with link quality of 55 or more In this phase stationary sensor node was configured as mesh router mode and battery powered sensor node was configured as an end node such that end node first communicated with intermediate router node and then gateway Thus the overall distance of transmission was 45 increased Full configuration of WSN in multi hop network topology is given in Figure 6 11 Gateway was connected to Laptop by using straight Ethernet cable bo 4 o se AN Fixed NI WSN 3202 AG Powered NI WSN 9791 AC Powered M
8. by adjacent AP This means nodes use the concept of hand over used in cellular technology 41 Mobility in WSN leads change in working environment and has more or less impact on WSN network architecture design working protocols range and power of transmission 7 3 Cost Size and Energy The cost size and energy of WSN depend on the application Some applications require microscopic sensor nodes whereas some require relatively big nodes with local processing and storing features Microscopic sensors are used in military applications They are cost effective small and consume less power but do not support local processing Local processing features on the nodes help to reduce communication overhead especially over low bandwidth links by pre processing extracting and transferring useful processed data rather than whole raw data 42 Big node with local processing and storing features are expensive and consume more energy 91 7 4 Communication Modality The common communication modality for WSN design is radio waves However there are numbers of other modalities such as diffuse light laser inductive and capacitive coupling and sound Most used modality is a protocol using ISM band Protocols using ISM bands are IEEE 802 11 IEEE 802 15 4 and IEEE 802 15 1 Zig bee IEEE 802 15 4 standard has a wide range and consumes less power so it is mostly used WSN communication modality 7 5 Infrastructure Designing sensor network is bas
9. can support Wi Fi b g standard The range of transmission is 30 meters and uses 24 Volt Direct Current VDC external power supply 17 Figure 3 1 shows the NI devices used for design I NI WLS 9163 with NI WLS 9234 I O Modules NI WAP 3701 3711 Figure 3 1 NI devices used for design I 3 1 1 2 Proposed Network Architecture When single WAP 3711 is configured it can communicate with sensor nodes WLS 9163 up to 30 meters range Omnidirectional antenna is used in such device therefore radiation pattern will be circle of 60 meters diameter approximately Figure 3 2 shows the radiation pattern of single WAP 3711 using omnidirectional antenna A E 9 WAP 3711 7 Q WLS 9163 60 Meters 2 Radio Link Figure 3 2 Radiation pattern of single WAP 3711 using omnidirectional antenna Advantage of omnidirectional antenna and WDS system are combined together to extend the coverage Maximum numbers of WDS that can be used are 6 18 Each WDS has overlapping distance of 5 meters such that first WAP 3711 has range of 55 meters and second third fourth fifth have range of 50 meters The last WAP 3711 has range of 25 meters Last WAP 3711 has 25 meters range as only one side can be used This last WAP 3711 is connected to the computer Figure 3 3 shows the overall connection diagram 20 WDS 1 WDS 2 WDS 3 WDS 4 WDS 5 WDS 6 lt a a al e 55 Meters 50 Meters 50 Meters 50 Meters 50 Meters 50 Meters 25 Meter
10. components 43 Different transmission protocols have different security types For example NI Wi Fi DAQ 9163 device can support up to the highest commercially available security IEEE 802 111 known as Wi Fi Protected Access 2 WPA2 Enterprises along with IEEE 802 1X authentication and Advanced Encryption Standard AES encryption algorithm However lower security features like Wired Equivalent Privacy WEP Wi Fi Protected Access WPA are also available for this device 44 Most sensitive Wi Fi WSN networks in military and industrial application uses advanced security management policies with at least one authentication server running a Remote Authentication Dial In User Service RADIUS Less sensitive Wi Fi WSN uses simple network security that means protocols could be WEP or WPA rather than IEEE 802 111 and encryption key could be Temporal Key Integrity Protocol TKIP rather than AES 7 10 2 Security Management Requirements Unsecured WSNs can suffer unauthorized access to the resource data and information Person or system attempting unauthorized access is called attackers and the process is called attack Managing WSNs security in order to avoid attack demands following requirements to be addressed 45 7 10 2 1 Authentication Authentication is the way to validate and identify user device or client of the networks Each device needs to be authenticated in order to use the resources Authentication helps a receiver to ensure dat
11. for the transmission of data WSN comprises all the advantages and disadvantages of wireless communicat10ns 4 1 Applications Major applications of WSN are described below a b c d e Home buildings and traffic security application Home and building surveillance for intrusion fire detection can be done easily using WSN Moreover WSN can be used in traffic surveillance technologies Data are updated constantly to provide historical and real time data of traffic count speed classification and re identification These data are then sent to Intelligent Transportation System ITS for the further processing 19 Automobile application In modern automobiles large number of sensors is installed for monitoring machineries conditions and health Large number of cables is used to interconnect these sensors with main cabinet These cables are replaced by using WSN technology By using WSN designer can reduce volume and weight of automobiles Industrial application WSN is used in industries to monitor temperature level and pressure of ovens pipes or tanks Furthermore health conditions of machineries can also be monitored For control purposes WSN is integrated with process management units Health monitoring application WSN is used as Body Area Network BAN 20 BAN consists of several sensors placed in different parts of human body These sensors measure heartbeat blood pressure neural activities etc
12. industrial process and pollution Varieties of sensors are available for underwater monitoring In most cases following measurement are made 1 Potential of Hydrogen pH 2 Dissolved Oxygen DO 3 Temperature 4 conductivity TDS 5 Turbidity pH measurement gives an idea of concentration of hydrogen ions in water DO measurement gives milligrams per liter mg l of oxygen availability in water Temperature measurement gives warmth or coldness of the water which in turn gives an idea about the amount of oxygen available Cold water can dissolve more oxygen Conductivity TDS gives an idea of the ion concentration Precisely it measures the ability of water to conduct the electricity Turbidity determines the cleanliness of water It gives a fair idea about how many particles are floating in the surface of water and how much sun light can reach beneath the surface of ocean and lakes 5 4 2 Network Architecture Basically there are two types of underwater WSN architecture 35 a Two dimensional b Three dimensional In two dimensional communication architecture the coverage area needs to be covered fully by sensor nodes and each node need to establish a multi hop path to the gateway In three dimensional architecture sensor nodes float at different depth of ocean in order to observe different seismic phenomena Sensor nodes are attached to buoy by means of wires and length of wires are adjusted as per requirement Two dimensional
13. nodes Measurement plot can be seen on the scopes This tab also consists of the current value of link quality temperature battery status Sensor node mode of configuration end node or mesh router node and presence of external power supply was indicted by Boolean indicators 43 Statistics tab as indicated in Figure 6 10 consists of scope for different parameters of temperature measurement like maximum value minimum value arithmetic mean and histogram for both sensors Numerical value for these measurements along with sample time for maximum and minimum value were also present in this tab Temp amp Link Qual Statistics Sensor Node 1 21 16 Plot 0 Link quality S1 100 Temperature S1 50 Plot 0 Current Link quality S1 Battery Voltage 51 oa 6 100 z Current Temperature_S1 752 Y o 2 i a 10 9 30 y T j z 0 50 D H rity Ted G 30 7 25 ES v 60 App a E a 0 0 20 Z 40 pr E 5 10 External Volatage SI Mesh Router 51 y gt dr Sensor Node 2 2244 PlotO tine qantiny 53 Plot 0 Temperature_S2 87 00 Current Link quality_S2 Battery Voltage_S2 100 50 E 40 Current Temperature_S2 Y ov i o 2 2 a 3 30 0 50 al 5 5 ne z U i a j 2 External Volatage_S2 Mesh Router_S2 e o I o a ty 3 3 Time Figure 6 9 Front panel consisting Temp amp Link Qual tab for both sensors Temp am
14. to underwater WSN node deployment 36 35 Figure 5 6 Two dimensional hierarchal topology using cluster concept ooooonnccncncncnnnnn 35 Figure 6 1 Full configuration in a star network topology rrrrrrrrnrrnnnnnnrrrrrvnvrvrnnnnnnerrrrnnvrrnnnnn 37 Figure 6 2 MAX window showing status of both sensor nodes detected by gateway 38 Figure 6 3 LabVIEW Project window showing both sensor nodes and its associated I O DN 39 Figure 6 4 Time vs Link quality graph to determine maximum distance of transmission 40 Figure 6 5 Top view of College to show where the sensor nodes and gateway were installed Figure 6 6 Connection diagram of PT 100 elements Transmitter and NI WSN 3202 41 Figure 6 7 Graph showing linear scaling to convert voltage to corresponding temperature 41 Figure 6 8 Block diagram of LabVIEW program used to measure temperature from 2 sensor INN 42 Figure 6 9 Front panel consisting Temp amp Link Qual tab for both sensors 44 Figure 6 10 Front panel consisting Statistics tab for both SENSOKTS oooooocnnnncnnnnncnnnnnnnnnnnnnos 44 Figure 6 11 Full configuration in a multi hop network topology ccccccccccoooocccnnnnnnnnonnnno 46 Figure 6 12 MAX window showing Network Mode for both sensor nodes where one sensor Node Is upaated as Mesh ROTE peers 46 Figure 6 13 Top view of College to show where the sensor nodes and ga
15. understand existing standard of WSN elements pros and cons of WSN case studies of different WSN applications and security management Two designs of WSNs were made using Wireless Fidelity Wi Fi and Zig bee Wireless Distribution System WDS concept was used in Wi Fi based design and router mode concept was used in Zig bee based design Zig bee based design was economical but supports the lower sampling rate Wi Fi based design was expensive but supports the high sampling rate up to 51 2 K samples per second per channel WSN was setup using NI Zig bee modules NI WSN 3202 sensor node and NI WSN 9791 Gateway were connected in star network topology and multi hop network topology Using multi hop topology indoor transmission range was increased significantly from 23 1 meters to 47 1 meters with link quality more than 55 Since the maximum sampling rate of Zig bee modules isl sample per second per channel physical quantity demanding high sampling rate are avoided in lab work In both topology cases temperature was measured Telemark University College accepts no responsibility for results and conclusions presented in this report Table of contents 1 6 INTRODUCTION 2d 10 Ml OBI CTIVES S s 11 12 REPORTS TRUC URD vr 11 STANDARD SCENARIO OF WSN seede 12 2ko STANDARD TRANSMISSION PROTOC OIDs pr 13 DEL DENN 13 DM ZN 14 EV 14 22 STANDARD NETWORK OPOLOG Y ido 15 Dod Weert Per NeIWOTK TOPILO SES 15 222 SIN Tole 16 2
16. with organisations and research groups working with similar problems in Norway and abroad Filename FMH606 WSN Saba Mylvaganam 6 rtf Signatures AD Kaye ge Student date and signature Supervisor date and signature VI O sa 01 02 67 Appendix 2 National Instruments price list National Instruments Norge Posiboks 177 1371 ASKER Tif nr 86 90 76 60 Fax or GG 90 76 61 H gskolen i Telemark er Asker 31 01 2011 Rabin Bilas Pant Kj lnes Ring 56 3918 PORSGRUNN Tilbudsnr 1235753 1 VI har gleden av lilby Dem f lgende Alt Antal Artikelnr Beskrivelse Listapels Rabatt Total pris 1 it 780997 11 NIWSN 3202 4 Ch 16 bit 410 Analog Input Node 3 009 10 00 2 7489 NOK International w 4 DIO channels Provides selectable input ranges and external sensor power up to 124 20m Leveringstid Ca 5 10 dager for lagerf rte artikler Dopirnalsesland HUNGARY 2 1 7980905 11 NIWSN 9791 WSN Ethernet Gatewey 9 30 powered 6 209 10 00 5 669 NOK International Version 9 20 DC powered Industrial temperatura and shock amp vibe ratings Outdoor range up to 1000m with line of sight Leveringstid Ca 5 10 dager for lagerf rte artikler Oppeinnalsesland HUNGARY 3 i 79047101 cAIO 9073 B Slot Integrated 266 MHz Real Time Controller 12 589 10 00 19 938 NOK and 2M Gate FPGA Leveringstid Ca 1 2 dager for lagerfarte artikler Oaprinnalsesland HUNGARY 4 1 781063 01 NI PS 15 Power Supply 24 VOC 6 A 10
17. 0 120 200 240 11669 10 00 1 529 NOK VAG Input Leveringstid Ca 2 5 dager for lagertarts artikler Coprinnalsesland China 5 to 7A050701 NI WLS 9234 IEEE 802 11b g Wireless Dynamic Signal 16 309 10 00 14 759 NOK Analyzer SW Selectable EPE amp AC DC Input 4 Ch 51 2 kch 2LBA 15 and NLDAGms Software Leveringstid Ca 12 20 dager for lagerf rte artikler D prinnalsesland HUNGARY SUM 36 086 NOK Fraki 164 NOK VAT 8 062 NOK Totalt 45 912 NOK Betalingsbetingelser 30 dager etter godkjent kredittsjekk Gyldig til og med 02 03 2011 Leveringstid Ca 2 3 dager for lagerf rte artikler eller 2 4 uker Se vedlagte salgsvilk r Neste side 68
18. 006 IEEE IAS Pulp and Paper Industries Conference USA 2006 27 Gangone M V Whelan M J Janoyan K D Deployment of a dense hybrid wireless sensing system for bridge assessment Structure and Infrastructure Engineering Maintenance Management Life Cycle Design and Performance 2011 7 369 378 28 Battista N D Wireless Monitoring the Longitudinal Movement of a Suspension Bridge Deck University of Sheffield UK 2010 29 Sukun K Pakazad S Culler D Demmel J Fenves G Glaser S Turon M Health Monitoring of Civil Infrastructures Using Wireless Sensor Networks Information Processing in Sensor Networks Apr 2007 254 263 30 Kim S Wireless Sensor Networks for Structural Health Monitoring University of California USA 2005 63 31 Melo M Taveras J Structural Health Moitoring of Golden Gate Bridge using Wireless Sensor Network Progress Report University of Massachusetts Lowel Aug 2009 32 DeltaV The DeltaV System Overviewed Emersion Process Management 2002 33 DeltaV DeltaV OPC NET Server Emersion Process Management Jan 2011 34 DeltaV Smart Wireless Gateway Emersion Process Management 2009 35 Pompili D Melodia T Akyildiz I F Deployment Analysis in Underwater Acoustic Wireless Sensor Networks Georgia Institute of Technology USA Sep 2006 36 Heidemann J Li Y Sayed A Wills J Ye W Underwater Sensor Networking Research Challenges and Potential Applications USC Information Science Institute USA
19. 25 MES ENEIWOTK L PJ 16 aid Iree NAWO k TN vasre 17 EXTENSION OF WSN COVERAGE sr 19 3A DESIGNING WSN FOR COVERAGE EXTENSI N gps 19 ILE Dent Use WE e e E E E E E E 19 Jd DESTE USA LEDO EO E E AE TA EA AE ENE ENA 22 313 Deston Selecion for Lab Work ia A R E E A A 24 APPLICATION ADVANTAGES AND DISADVANTAGES OF WGN cccsssssseccccccessssceeseeeees 25 APPEICA TON ee 25 AD NN eee 26 AD DISADVANTAGES sas ses trade tacwanensstadguscosanchieinadabincmaneeslsiadeuseesnnss hasten Satea eye oige a boolean clio iria 26 22 INDUSTRIAL APPLICATIONS ISSUES A A NE a EE 2 CASESTUD cua A A A A A A AAA aane E 29 CAS STUD Li ds li 29 We 21016 TONG RE SEE EE REE EE RE 29 TD Svene 29 I ACASE AA e rn usadwabedlahendeudacmoiebiesdastuaackars 30 VIL BENN 30 ele NNM NES 31 525 Damn OS San 31 A EN 32 IoT BA REE EE 32 M32 Existing Process Management Systemi arve 32 Do COUPE EE SEE N A 33 CASE STUD IV iaa a E A E AS 34 SAd BR MN 34 542 NECIOS ARCE CUE seede 34 OD Cms STAN SN AAA AE AAA AA AO 36 LAB WORE aiii de 37 6 1 PHASE l STAR NETWORK TOPOLOGY cdi 37 OGLA oe 37 6 1 2 Determining Maximum Distance of Transmission rrrrrrrrrnnrnnnnnnnnrrrnnnnnnnnnnnrrnnrnnnnnnnnrrnnrrnnnnnennsnnne 39 OLD Frere Ser Uta 41 od VEN COOP 42 odd FP onPmdama Data NEPOS 43 62 PHASE I MULTI HOP NETWORK TOPOLOGY gassen 45 6 2 1 Determining Operation of Mesh Router Mode rrrrrnnnnnnnrrnnrnnnnnnnnrrnnnnnnnnnnnrnrnn
20. II This study is related to the work done to measure and monitor a physical dynamics like vibration and displacement in order to determine the structural health condition of a bridge 5 2 1 Background Structural condition of a bridge is affected by different factors like external forces wind seismic activity and traffic passing through it An application of WSN is to monitor structural health conditions in order to know how structures perform over a long period under changing environmental conditions Dense wireless sensor solutions as used in suspension bridge at St Lawrence County New York 27 are practiced in many cases of structural monitoring In Dense wireless sensor solutions large numbers of sensor nodes are distributed throughout the bridge structure Factors such as strain vibration acceleration and displacement measurement are analyzed at the same time However there are also many cases where single nodes can be used to monitor particular location of a bridge The University of Sheffield succeeded in monitoring longitudinal movement of Tamar Bridge using single measurement node 28 30 5 2 2 Network Architecture WSN deployed in Golden Gate Bridge USA for structural health monitoring is similar to the concept of dense wireless sensor solutions 64 sensor nodes were installed on 4200 ft long main span and the tower located at the southern end of the bridge Out of 64 nodes 46 nodes were made hop network for routing da
21. Master s Thesis 2011 Candidate Rabin Bilas Pant Title Wireless Sensor Networking with Lab Scale Intermediate Measurement Node for Extension of WSN Coverage Telemark University College Faculty of Technology Kj lnes 3914 Porsgrunn Norway Lower Degree Programmes M Sc Programmes Ph D Programmes TFver 0 9 gt 3 y 7 TODA ETE Telemark University College Faculty of Technology M Sc Programme MASTER S THESIS COURSE CODE FMH606 Student Rabin Bilas Pant Thesis title Wireless Sensor Networking with Lab Scale Intermediate Node for Extension of WSN Coverage SISNMUTES 9 parecia dra arias dnd Number of pages 68 Keywords WSN Router Coverage Sensor Node Supervisor Saba Mylvaganam SONO NS 2 Supervisor Hans Petter Halvorsen SIME hawt thc ec ea Censor te External partner Frode Skulbru NI Norway SONS Leda EE EE EN Availability Open Archive approval supervisor signature Sign o ooooooooooooooooo o EE Abstract Wireless Sensor Networking WSN will be leading solution for data communication for various monitoring and control applications in industries However the technology is not widely implemented due to some limitations Improvement in transmission range security issues real time monitoring and control issues system integration and coexistence are the fields in WSN which require more investigations In this thesis a comprehensive literature survey was done to
22. Similarly link quality of sensor node 2 fluctuated from 83 to 95 The reason behind link quality fluctuation was due to opening and closing of a door as people entered the room The current temperature current battery voltage current link quality for sensor node 2 were measured to be 22 4 C 6 V 87 respectively External Voltage Boolean indicator was highlighted which indicated as sensor node2 was powered by external AC power supply but Mesh router Boolean indicator was not highlighted that means sensor node 2 was configured in end node mode not at the mesh router Temperature measurement for senor node 1 fluctuated constantly around 20 9 C to 23 3 C Link quality fluctuated from 55 to 65 The current temperature current battery voltage current link quality for sensor node 1 were measured to be 21 1 C 0 V and 62 respectively Both the Boolean indicators for sensor node were not highlighted that means there was no external power supply to node and node was configured in end node mode not mesh router mode From Figure 6 10 statistical interpretations for both sensor nodes were studied It can be seen that the maximum temperature for sensor node 2 was measured to be 23 1 C at ig sample time and minimum temperature was measured to be 18 C at 337 sample time Mean value of temperature was found to be 20 63 C Histogram for sensor node 2 suggested that temperature around 18 C occurred most of the time and temperature
23. WSN design fails Some solutions can be implemented in order to measure the physical parameters with a high sampling rates They can be followings a Various forms of spatial and temporal compression of data can be done before transmitting These compressed data are recovered in receiving sections b Temporary storing of data can be done in local nodes and allowed to transmit over network when bandwidth becomes free 7 10 Security WSN uses RF signal as a physical transmission medium A wireless medium adds security challenges than that of wired system Due to this sensitive data needs to be protected from unauthorized access There are many common security practices which are based on network security components like wireless security protocols encryption authentication etc 7 10 1 Security Management Management of security is done by incorporating appropriate security components as per the nature of networks public private W Fi Zig bee Figure 7 1 shows the secured WSN architecture This architecture comprises of WSN Network Manager and Security Manager AP or Gateway is connected to Network Manager and Security Manager Security Manager plays a vital role to maintain secured networks by authenticating network devices and by generating storing and managing encryption keys 53 op O Gateway Acess Point AD Network Manager Security Manager amp Sensor Node Figure 7 1 WSN Architecture with security
24. Zig bee WSN using sensor node as router 23 3 1 2 3 Cost Estimation Device going to be implement in second design their individual cost and grand total after VAT and discount is given in the Table 3 2 Table 3 2 Cost estimation of WSN using NI Zig bee Modules Cost Estimation for Design2 Zigbee Network using NI devices Devices Price per unit NOK Total price NOK NI WSN 9791 gateway 5669 00 1 5669 00 NI PS 15 Power Supply 5 A 24 VDC 1529 00 3058 00 Sum 14305 00 VAT 3826 Grand Total 18131 00 Total cost for design based on Zig bee 1s approximately 18131 00 NOK This cost has been calculated with the reference to price list provided by National Instruments which is included in Appendix 2 3 1 3 Design Selection for Lab Work Cost for Design 2 is very less as compared to Design 1 Design 2 is best suited in the context where low sample data rate are required to be measured Design 2 can be easily modified to mesh network topology but this is not possible in Design 1 Apart from these Design 2 consists of less number of devices giving advantages of less installation time less labor reduced complexity etc Due to these reasons Design 2 seems to be best option for a lab Work 24 4 Application Advantages and Disadvantages of WSN WSN has various applications in home industries health sector and environmental monitoring There are also some merits and demerits of WSN Since radio medium is used
25. a and control information are originated from valid source 7 10 2 2 Encryption Process to modify data or information such that it can be read or identified by intended user device or client is called encryption Strong encrypted keys and effective key management scheme are used to encrypt data prior to transmission At the receiving section receiver needs to know the proper decrypted keys to decipher the encrypted message 54 7 10 2 3 Integrity and Data Freshness Integrity refers to the change of data before it reaches to an intended receiver Data change may occur due to communication environment faulty nodes or intended attack by attackers Data s freshness ensures that receiver 1s receiving new data in each cycle and no old data are replayed at the receiver To ensure data freshness time stamp is added in each packet of data Other requirements that are equally important for security management are listed below a Sensor nodes are capable of avoiding the intentional jamming that makes a sensor unavailable b Sensor nodes are capable of self organizing and self healing in different adverse environment and attack c Location information 1s core part of WSN Unsecure localization scheme in WSN allows an attacker to manipulate false signal strength 7 10 3 Protocol Stacks Security Threats and Prevention Protocol is the set of rules used by communication devices to exchange data In WSN sensor nodes used layer based protoco
26. achieved 8 4 Security In order to use WSN in application like industrial control and monitoring military purposes etc network requires minimum level of security to avoid attacks Data needs to be sufficiently encrypted properly authenticated and any change or replay of data during transmission needs to be identified and stopped Jamming tampering collisions unfairness misdirection misinformation flooding de synchronization were identified as different possible attacks when referred to ISO OSI protocol Jamming and tampering are physical layer attack and can be avoided by spread spectrum technique and tampering free packaging Data link layer attacks like collision and unfairness can be prevented using collision detection techniques and making small frame of data such that they occupy channels for very less time Network layer attacks like packet dropping and misrouting can be prevented by encryption authentication multi path routing and use of unique key Flooding and de synchronization are identified as major attacks in the transport layer They are prevented by client puzzle techniques and proper authentication techniques 99 9 Conclusion and Future work After achieving thesis requirements to certain level important conclusions were drawn out Apart this work also gives ideas about how this thesis can be used in future in more elaborated and innovated form 9 1 Conclusion The following conclusions are drawn f
27. al indicators and waveform graph The output of Link Quality variable is between I to 100 this output is shown by numerical indicatiors and waveform graph This shows how good is the the air link quality between sensor nodes and gateway The output of Battery Voltage variable is between 1 to 6 this output is shown by numerical indicatiors I means very less battery volta e condition The output of External Power variable is either TRUE or False this output is shown by bollean indicators If TRUE sensor Histogram of temperature measuremt throughout the monitored time period Histogram is plotted on waveform chart Data in each loop are stored in 1 Dimensional array and then passed to next loop by using Shift register Inbuilt function Statistic Express VI is used to calculate mean vlaue maximum value minimum value of temperature throughout the monitored time period Furthermore sample time when minimum value and maximum value occured is also calculated by this VI These values are shown by numerical indicators and waveform chart Merge signal function is used to concatinate and show all data in single graph Data in each ae loop are stored in 1 Dimensional array and then passed to E Shift reg ter 6 1 5 Front Panel and Data Interpretation Front panel of LabVIEW program was divided into two tab control Temp amp Link Qual tab as shown in Figure 6 9 consists of scope for temperature and link quality measurement for both sensor
28. architecture may not observe these 34 phenomena as sensors are installed at the bottom of sea level Figure 5 5 gives one possible approach to underwater WSN installation al nt AAA a ee Figure 5 5 Possible approach to underwater WSN node deployment 36 Large numbers of sensor nodes are installed at the bottom of the sea level These nodes are battery powered and remain in sleep mode in order to save power At the top layer control nodes are installed in offshore or onshore platform For large area coverage super nodes are installed Super nodes have access to lower level sensor nodes and upper level base stations 36 In underwater WSN hierarchical topologies are inevitable as nodes are deployed in the bottom of sea level 37 For this multiple sink nodes AP are installed Each AP can talk only with the group of sensor nodes called cluster and super nodes Figure 5 6 is a hierarchical topology developed after combining two dimensional architecture and cluster concept shown by dotted line Figure 5 6 Two dimensional hierarchal topology using cluster concept 35 5 4 3 Comparisons with Terrestrial WSN Underwater WSN is more problematic than terrestrial WSN in terms of radio propagation fault battery backup and signal attenuation 38 Comparisons of underwater WSN with terrestrial WSN are listed below a Radio communications inside water always suffers from long propagation delay fading and mu
29. ase I of lab work was modified with the provision of measuring temperature statistics only for end node Since there were no significant variations in temperature histogram analysis was removed Code still has provisioned of link quality monitoring battery voltage monitoring along with Boolean indicators for external voltage and 48 mesh router mode for both nodes Figure 6 17 shows the code and Figure 6 18 shows the front panel of LabVIEW program used in this phase of lab work Temperature 51 Link quality 51 End node hd Eser ES Link quality 51 Ho o Mean Value 51 AT m Max Value 51 a e Battery Voltage s Statistics e Signals Time of Maximum 51 ETE dr Arithmetic Mear l pin Value s1 Maximum pp Soe at Mesh Router ETETE E Sanhita P HJEL f Time of Maximo jj Minimum pt Time of Minimum SI Time of Minimu oa EDEL soe Link lity 52 Rout de 2 gt a Statistics 51 ink quality 52 Router node om 0 bo Escr Sy das IR a ie Link Quality gt a e Battery Voltage s External Volatage 52 d we External Power stop ee Mesh Router 52 a a Me Mesh Router 1000 Figure 6 17 Block diagram of LabVIEW program used to measure temperature from end node Sensor node_ S1 End node Senosr node_ 52 Router node z E _ er 2377 Plot 0 AN i Statistics SI Arith AN Maxim YY Minim AA i 50 50 Link quality S1 Battery Voltage 51 D T 100 6 o 40 a 40 52 Ext
30. ation sensors was installed on the top of the tower Second sensor node with temperature sensor was placed on one corner of the floor and third sensor node was placed on the corridor Figure 5 4 Engineers busy on random installation of wireless HART in Yara Porsgrunn First of all Engineers tried to form a star network topology During testing it was found that gateway could not receive signal form any of the sensors installed This could be because of 33 obstacles like walls of the buildings and huge machines causing diffraction or deep fading of the radio signals Due to this Engineers decided to put a sensor node in between the gateway and measurement points such that multi hop network is formed and signals can be received in the gateway This approach was found successful Later on complete mesh topology was formed by installing multiple intermediate sensor nodes and measuring nodes 5 4 Case study IV In this case study underwater WSN for water quality monitoring has been studied Under water sensor networks have wide applications in industries They help in water quality monitoring ocean graphic data collection disaster detection amp prevention and oiled field monitoring 5 4 1 Background To control physical chemical and biological characteristics of water water quality monitoring is required Industries found it very important because it helps to check contaminations of water discharge of toxic chemical from
31. controller Controller monitors the process information and takes necessary action by allowing actuators to act Traditionally these functions are realized in a wired system using Field bus and Profibus 24 These days WSN based non intrusive methods are also practiced 5 1 1 Background In industries motor driven systems consume most of the energy used Research reveals that 98 of motors used in industries are below 200 hp and consume 85 of overall energy used in industries 25 This adds the importance of study on industrial plant energy evaluation and planning In many cases motors do not operate on full load condition which results in the reduction of efficiency and waste of energy Average load consumed by motors in industries is 60 of rated capacity 26 Moreover motor condition monitoring helps to prevent unexpected motor shut downs decrease energy consumption and increase efficiency This information gives the plant manager valuable information for planning a scheduled maintenance WSN based close loop industrial plant energy evaluation and planning system uses non intrusive methods for efficiency estimation This WSN transfer only motor terminal voltages and currents 5 1 2 System Architecture The system architecture is divided into two parts Motor Control Centers MCC s and Central Supervisory Station CSS Terminal qualities measurements are done at MCC s and transferred to CSS through WSN These data are collected a
32. d array were used to store and process the values in every loop in order to calculate arithmetic mean maximum value time of maximum value minimum value time of minimum value and histogram Temperature 51 gt gt Current Temperature 51 j Signal Histogram 1 mer S d gj Link quality_51 Histogram lt H PA A we Mean Value 51 DEL Max Value 51 i Statistics P E Signals if External Volatage_51 i error in no error p z HH n E External Power it Mesh Router 51 I TE or Maximum pg ff 22 Mesh Router f Temperature 52 SSS Current Temperature 52 4 F F Create Histogram iH ER so Histogram Link quality 52 Histogram e SSS a hj Ella Mean Value 52 Max Value 52 AR Battery Voltage f External Volatage 52 Ek Signals ke 1 cui amp 4 3 External Power error in no erro j if Mesh Router 52 Arithmetic Meare 2 Mesh Router Tab Control FME Bi Statistics 52 e O Era rer Statistics 1 Le EE Time of Minimum 52 Figure 6 8 Block diagram of LabVIEW program used to measure temperature from 2 sensor nodes 42 Table 6 1 Elements used in LabVIEW code and their functions Output of AIO is 1V 5V linear scaling is done to convert voltage measuremet to corresponding temperature measurement I V corresponds to 0 deg Cent and SV corresponds to 50 deg cent Temperature measurement is shown by numeric
33. ed on the infrastructure used Sensor nodes can communicate with each other either using infrastructure or without using infrastructure for e g sensor node NI WLS 9163 can be connected either using infrastructure like NI WAP 3711 or using ad hoc mode with computer Infrastructure is simple AP or gateways installed in industries schools and public areas 7 6 Network Topology and Coverage WSN using single infrastructure are considered to be in star network topology where coverage for WSN depends entirely on its infrastructure coverage area In a multi hop network the coverage for WSN depends on the sum of range supported by their sensor nodes and their spatial distributions Using multi hop the network topologies that can be created are mainly mesh extended star and tree Moreover coverage of WSN depends upon output power of the transmitter Different countries have different regulations defined for maximum output RF power Output power of ISM band ranges from 0 dBm to 20 dBm 8 7 7 Network Size Network size of WSN varies from few nodes to numbers of sensor nodes The number of sensor nodes in particular WSN depends upon interest of measurement quality of WSN environment and coverage area WSN with fewer sensor nodes is flexible to relocate to manage traffic to upgrade and troubleshoot Implementing WSN with large numbers of sensor nodes demands more research in designing communication protocols and routing algorithms 7 8
34. ement 1 V to 5 V alg NI WSN ai 0 3202 GND ai 1 ai 1 GND GN E DC 24VDC Figure 6 6 Connection diagram of PT 100 elements Transmitter and NI WSN 3202 Transmitter operating range is O to 50 C Due to this limitation any temperature below and above this range cannot be measured The voltage drop between resistor points was always I to 5 V I V corresponds to 0 C and 5V corresponds to 50 C A linear scaling was done in a LabVIEW program in order to convert voltage measurement to corresponding temperature measurement Figure 6 7 shows a linear scaling graph for the corresponding temperature measurement at the different voltage level 50 C Figure 6 7 Graph showing linear scaling to convert voltage to corresponding temperature 41 6 1 4 LabVIEW Code The program code was made in a block diagram as shown in Figure 6 8 The code was written in a while loop All the interested I O variables were dragged and dropped to the block diagram from a LabVIEW project window Indicators for each variable were created Some of the important elements of code and their functions are given in Table 6 1 Voltage measurement from sensor nodes was converted to temperature measurement by linear scaling For signal analysis Create Histogram Express VI and Statistic Express VI were used Indicator and Scope for each parameter were created Shift register and Buil
35. emp lvproj a g H i Dependencies r Build Specifications E NI WSN9791 01563E3C 169 254 62 215 E Nodel ID1 1598843 NI WSN 3202 i RR AT AT nan AR Battery Voltage DICO DIO1 DIO2 DIOS External Power Link Quality Mesh Router fe a VI Deployed Node ID 2 1598845 NI WSN 3202 ADD mnt AM AL ai AR Battery Voltage DIGO DIO1 DIO2 DIOS External Power Link Quality Mesh Router VI Deployed Figure 6 3 LabVIEW Project window showing both sensor nodes and its associated I O variables 16 New VI was created I O variables were dragged and dropped to the block diagram and indicators for each I O variables were created 6 1 2 Determining Maximum Distance of Transmission After the coupling and configuration of sensor nodes with gateway second step was to determine maximum wireless distance that WSN devices can support without traffic interruption For this sensor node powered with battery was made mobile Mobile sensor node was slowly taken away from the gateway and link quality was monitored in LabVIEW program 55 of link quality was supposed to be minimum signal strength as instructed by National Instrumens and regarded as a fair signal that corresponds to signal strength of 2 bar highlight in the sensor nodes 39 39 sensor node was moved away further up to the distance where signal strength decreased to 60 Measurement of link quality was done in LabVIEW simultaneously When sig
36. er Network Topology 2 2 2 Star Network Topology Star network topology is a technology in which numbers of remote nodes can send or receive data to single base station Unlike peer to peer network nodes are not permitted to send and receive message to and from each other Routing algorithm for this topology is simpler than other topologies There are some disadvantages also first remote nodes must be in radio range of base station Second due to dependability of all remote nodes on a single base station redundancy and robust structure are hard to achieve Base station 1s core part of this network as it is responsible for message handling routing and decision making Figure 2 3 shows how 3 sensor nodes are connected to single base station in star network topology O Base Station Access Point Sensor Node Radio Network Figure 2 3 Star Network Topology 2 2 3 Mesh Network Topology Mesh networks are distributed networks which allow any node in the network to talk to any other nodes in the same network within its coverage area Mesh network can be suitable for large scale distributed network of sensors over geographic region like personal or vehicle security surveillance systems 12 Mesh network topology has the advantages of fault tolerance and load balancing and disadvantages of scalability 13 For an instance if an individual node fails a remote node still can forward message to desired node by communicating to any
37. ernal Volatage_S1 dh HO 30 3 30 50 3 Y y z E 5 5 25 2 9 Mesh Router S1 20 5 20 19 E E Max Value SI Min Value SI Mean Value S1 o 24 8373 23 7673 24 2373 Time of Maximum SL Time of Minimum S1 90 890 Link quality S1 End node 67 00 Plot 0 AA i Link quality 2 Router node 2 65 00 Plot 0 AN i Link quality 52 Battery Voltage 52 100 6 F External Volatage_S2 Be 5 J E 50 3 3 z E 2 0 Router 52 E 0 0 l I l i li I li 0 200 400 600 800 1023 0 200 400 600 800 1023 STOP Time A Time Figure 6 18 Front panel consisting temperature statistics for end node and link quality statistics for both end node and router node 49 Mesh router boolean indicatror for sensor node 2 was highlighted indicating the operation of router mode and that of sensor node I was not highlighted indicating the operation of the end node Link quality of both nodes were more than 55 indicating fair link quality Statistical analysis for end node showed that maximum environmental temperature was 24 8 C at 90 sample time minimum temperature was 23 7 C at the 890 sample time and average temperature throughout observation duration was found to be 24 2 C Current link quality for end node and 2 were 67 and 65 respectively Boolean indicator for external voltages indicates that router node was operated by external power supply Battery voltage of the end node was f
38. es were detected File Edit View Tools Help a Y My System C Restart bd Ra Refresh Add Node X kE gt O gl Data Neighborhood ag Devices and Interfaces 44 Scales Wireless Channel 14 Y El Software IVI Drivers YI Node Type Serial Number ID Last Communication Time Battery State Link Quality Network Mode Firmware Version a amp Remote Systems i F NI WSN9791 01563E3C S NIWSN 3202 1598843 1 01 01 1904 03 12 08 OK Excellent End Node wsnVoltageNode 1 0 0f0 Firmware NI WSN 3202 1598848 2 01 01 1904 03 12 08 No Battery Excellent End Node wsnVoltageNode 1 0 0f0 Firmware TT ES System Settings bye Network Settings WSN Nodes Gateway Firmware Connected Running Figure 6 2 MAX window showing status of both sensor nodes detected by gateway 10 11 pA 13 14 15 After that devices were configured in LabVIEW New LabVIEW project was opened Project name was right clicked and New gt gt Targets amp Devices were selected Existing target or device was selected and WSN gateway folder was expanded After certain scanned time NI WSN 9791 was detected OK button was clicked so that gateway was added to the Project Explorer window In order to see associated nodes NI WSN 9791 was expanded Each node was expanded and all I O variables were seen Screen shot of LabVIEW project and its associated I O variables for both sensor nodes were given in Figure 6 3 38 E bd Project testt
39. esh network can provide redundancy to network Another way for coverage extension is the use of multiple Access Point AP to create Wireless Distribution System WDS In WDS system each AP can communicate with adjacent APs and its associated sensor nodes 1 1 Objectives The main objective of this thesis is to design and implement WSN in order to increase the transmission range by using intermediate router nodes that can communicate with sensor node and gateway In order to achieve this following sub objectives are considered a Study of current standard scenario of WSN in terms of network topology and transmission protocol b Different applications pros and cons of WSN Deployment challenges of WSN in an industrial environment c Four case studies for different applications of WSN d Implementation issues of WSN security threats and management e Lab work to verify improvement of transmission distance f Finally documentations of lab result and theory The task description is given in Appendix 1 1 2 Report structure This report consists of nine chapters Chapter one will give general introduction of WSN and objectives of a thesis Chapter two will discuss about existing standard scenario of WSN in terms of transmission protocol and network topology Chapter three will discuss about W1 Fi and Zig bee based design for coverage extension Applications advantages and disadvantages of WSN will be discussed in chapter four Fo
40. f transmission depends upon the types IEEE 802 11 that are used by Wi Fi devices Five types IEEE 802 11 standard are 802 11 802 11a 802 11b 802 11g 802 11n Their ranges vary from 30m to 125m and data rate varies from 2 Mbps to 54 Mbps The newest technology 802 11n supposed to have data rate up to 100 Mbps to 200 Mbps 11 Modulation scheme frequency band range and data rate are given in Table 2 3 Table 2 3 Wi Fi specifications Range Bit Rate Band Types Modulation meter Mbps GHz IEEE 802 1 1 g 20 t0 54 OFDM DSSS IEEE 802 11 n 125 100 to 200 OFDM 2 2 Standard Network Topology Network topology defines the arrangement and connection of computers or nodes with each other WSN network topology describes how sensor nodes are connected to other sensor nodes or hubs or base stations Computer topologies include Star Ring Fully Connected Bus Tree and Mesh 12 This chapter focuses mainly on following four standard WSN data network topologies a Peer to peer Network Topology b Star Network single point to multipoint Topology c Mesh Network Topology d Tree Network Topology 2 2 1 Peer to Peer Network Topology In this type of topology each nodes can communicate directly with another node without usage of any centralized infrastructure or hub Figure 2 2 shows how three sensor nodes are connected on peer to peer network topology Sensor Node Radio Network Figure 2 2 Peer to Pe
41. h nodes were communicating directly with gateway with link quality ranges from 55 to 65 Reset button of an end node was pressed for more than 5 seconds and released After this end node searched for the strongest link nearby Since the nearest link would be the waves 46 propagated by router node it was connected to router node rather than the gateway The dotted line in Figure 6 13 shows the new connection path al E New path followed by end node after reseting Figure 6 13 Top view of College to show where the sensor nodes and gateway were installed and new path followed by end node after resetting Link quality of the end node was measured in LabVIEW throughout the testing period Figure 6 14 shows the link quality of the end node before and after 1t has been reset At first the link quality was seen to be 55 to 60 After resetting 1t was observed that end node link quality increased sharply and reached to almost 100 This was because of a new link that has been established between the end node and router node sa New link with Router Aa pan LI i Link quality 25 Time Figure 6 14 Time vs Link quality graph to observe improvement in link quality after end node has been reset 47 significant change in MAX window at this point was observed Fair link quality of the end node as in Figure 6 12 was changed to Excellent This also proved the improvement in link quality after t
42. he end node was reset Figure 6 15 shows the change in MAX window when the end node was reset File Edit View Tools Help 2 9 My System 1 Restart bl cs EY Refresh 42 Add Node X Remove Node Configure Node opp Update Firmware Cancel gt 8 Data Neighborhood F b a Devices and Interfaces b 44 Scales Wireless Channel 14 v b amp Software gt E M Drivers a amp Remote Systems Node Type Serial Number ID Last Communication Time Battery State Link Quality Network Mode Firmware Version amp NIWSN 3202 1598843 1 01 01 190404 10 08 OK C Excellent 3 EndNode NI WSN 3202 1 2 0f0 amp NIWSN 3202 159884 2 01 01 190404 10 08 NoBattery Far Mesh Router wsnVoltageNode 1 0 0f0 Firmware Figure 6 15 MAX window showing Excellent link quality after the end node was reset Taking advantages of improved link quality end node was moved away from router node until its link quality reached around 55 Thus overall distance of transmission was increased Figure 6 16 shows the location of the end node and router node and their distance from gateway The end node was kept 24 meters away from router node Thus total transmission range was measured to be 47 1 meters o pr n image DD 2011 DigitalGlobe r 2011 Google Figure 6 16 Top view of College to show where the sensor nodes and gateway were installed and their separation distance 6 2 2 LabVIEW Programming and Data Interpretation LabVIEW code used for ph
43. ic J A Denial of Service in Sensor Networks Computer Oct 2002 35 54 62 64 48 Nationsl Instruments CompactRIO Third Party Modules National Instruments Corporations USA Dec 2010 49 SEA Zigbee http www sea gmbh com en products compactrio products sea crio Modules wireless technology zigbee Apr 2011 65 11 Appendices Appendix 1 Task description S 3 ST MONA A Telemark University College Faculty of Technology FMH606 Master s Thesis Title Wireless Sensor Networking with lab scale intermediate measurement node for extension of WSN Coverage TUC supervisor Saba Mylvaganam Hans p Halvorsen Frode Skulbru NI External partner National Instruments Task description This study will have focus on wireless sensor network However to facilitate the implementation of wireless networking an overview of existing standards in wireless networking will be essential The topics to be addressed for an in depth study and demonstration are 1 Overview of the current standards scenario with essential technical details 2 Pros and cons of wireless networking and especially wireless sensor networking in the industries 3 At least four case studies of wireless sensor networking as practised in the industries 4 Lab scale demo unit consisting of two wireless nodes communicating with an intermediate measurement node as a router 5 Identifying and describing issues related to implementation of wireless sensor
44. ing and managing central hub in common central point Simple example of tree network topology is given in Figure 2 7 000 Figure 2 7 Tree Network Topology Central Hub Root Node Sensor Node Radio Network 3 Extension of WSN coverage The range between source sensor node and sink monitoring node depends upon transmission protocol used Zig bee protocol offers highest transmission range at the cost of lower sampling rate Zig bee devices have coverage up to 300 meters Wi Fi devices have coverage up to 30 meters 802 I 1 n is newest technology with 125 meters range One way to increase range 1s to use high power directional antennas However different countries policies to limit the maximum output power in ISM bands restrict the use of high power directional antennas Maximum output power limitation in USA is 50mW Thus the maximum range supported by Zig bee is 300 meters For Europe and Asia maximum output limitation is I 0mW and the corresponding maximum range supported is 150 meters 14 Another way to increase the range is to place infrastructure in between source and sink nodes This infrastructure acts as a repeater Basically AP can be used as an intermediate infrastructure in Wi Fi based WSN whereas router can be used in Zig bee based WSN These two methods are studied and technical overview is presented in the following sub chapters 3 1 Designing WSN for Coverage Extension In order to design the extens
45. instructions and specifications NI 9234 National Instruments Corporations USA Aug 2008 62 17 National Instruments NI WAP 3701 3711 User Manual National Instruments Corporations USA Sep 2007 18 National Instruments Wireless Data Acquisition Range versus Throughput National Instruments Corporations USA Jun 2007 19 Cheung S Y Varaiya P Traffic Surveillance by Wireless Sensor Networks Final Report University of California USA Jan 2007 20 Tavares J Velez F J Ferro J M Applicaiton of Wireless Sensor Networks to Automobiles Measurement Science Review 2008 8 65 70 21 Garcia L R Lunadei L Barreiro P Robla J I A Review of Wireless Sensor Technologies and Application in Agriculture and Food Industry State of The Art and Current Trends Sensor 2009 9 4728 4750 22 Paavola M Wireless Technologies in Process Automatation Review and an Application Example University of Oulu Finland Dec 2007 23 Wireless HART IEC 62591 WirelessHART System Engineering Guide Emerson Process Management Revision 2 Oct 2010 24 Khakpour K Shenassa M H Industrial Control using Wireless Sensor Networks K N Toosi University of Technology Iran 2007 25 Department of Energy Industrial Wireless Efficiency amp Renewable Energy Report USA Dec 2002 26 Gutierrez J A Durocher D B Lu B Harley R G Habetler T G Applying wireless Sensor Network in Industrial Plant Energy Evaluation and Planning Systems The 2
46. ion of WSN coverage task is defined at first The task is to measure any physical quantity which is placed more than 300 meters far from monitoring station The overall concept of these designs will be to study all the technical constraints and budget requirements Comparisons between the designs will be made and one design will be chosen for lab work 3 1 1 Design 1 Using Wi Fi This design is based on the concept of WDS WDS is a system where new AP is placed between existing sensor node and AP By doing this the overall range of the system 1s increased approximately by twice More APs can be used to create more WDS such that each intermediate AP acts as repeater Many Wi Fi vendors are using IEEE 802 1 1a b g multimode AP IEEE 802 11 b g mode is mainly used to connect sink AP with computer or monitoring station IEEE 802 11 a mode is mainly used to connect two APs in order to form WDS 15 In this case for both purposes the device supporting IEEE 802 11 b g multimode will be used 3 1 1 1 Technical Specification of Device Used Proposed devices for design using Wi Fi are from National Instruments NI WLS 9163 C series carrier along with NI WLS 9234 I O module is used as sensor nodes This device supports 4 input channels with 24 bit resolution and maximum sampling rate of 51 2 K 19 samples per second per channel It requires 220 Volt Alternative Current VAC power supply 16 NI WAP 3701 3711 is used as access point which
47. jacent channel interference multipath propagation reflection scattering and diffractions There are potential chances to degrade reliability of data by noise generated by heavy equipment by multipath propagation 22 and by interference by other devices using ISM bands 23 There are two types of interference broadband interference and narrowband interference Table 4 2 summarizes all possible interference sources that can exist in industrial environments Table 4 2 Source for Interference 22 Broadband interference Narrowband Interference Inverters Silicon Control Rectifier SCR Radio and TV transmitter circuits Computer ESD Signal generator Lightining electromangnetic pulses Microwave and ultrasonic equipments 27 Interference caused by other devices using ISM band gives co existence problem Wireless HART use many methods to avoid co existence problem 23 These methods are a Network Segmentation b Spectrum isolation c Low power d Spacial hopping e Channel hopping f Direct sequence spread spectrum coding g Time synchronized mesh protocol 28 5 Case Study Four case studies related to different industrial applications of WSN were studied and presented in this chapter 5 1 Case Study I In this case study WSN based closed loop industrial plant energy evaluation and planning system has been studied In a closed loop system a sensor collects status information of a process and feeds it to a
48. king hole during wire installation 4 3 Disadvantages Some disadvantages of WSN are given below a b c d WSN provides low speed of communications as compared to wired sensor network In some cases hackers can hack the information if proper security policies are not implemented on the networks Most of WSN uses ISM band and vendors for multiple applications also use the same spectrum In such case there exists problem of co existence due to interference Some sensor nodes are required to be installed in the remote places and unattended for number of years These nodes are powered by battery So there may be power problem after some years of deployment 26 e WSN are relatively more complex to configure than that of wired system and there 1s always problem of bandwidth for special measurement cases Abovementioned advantages and disadvantages of WSN are summarized and presented in Table 4 1 Table 4 1 Advantages and Disadvantages of WSN Advantages Disadvantages Avoids wires Low speed compared to wired network Low cost small size and less power ti pe Security issues less transmission range consumption Easy installation compare to wired network Coexistance problem Relocation and restructuring is easy Battery drained after long use Avoids drilling in walls Bandwidth problem 4 4 Industrial Applications Issues WSN operating in industrial applications may suffer from co channel interference ad
49. l stack based on International Organization for Standarization ISO Open System Interconnection OSI protocol 46 However there are lots of researches going on to determine perfect protocol stack for WSN Sensor node protocol stack based on ISO OSI protocol is given in Figure 7 2 Wood and Stankovic conducted research to outline the possible attacks and prevention in first 4 layers of the protocol stack 47 Transport Layer Network Layer Data Link Layer Physical Layer SUE q juswabeuey Aygo SUELA JUSULIS EUEIN SE Figure 7 2 Sensor node protocol stack 46 7 10 3 1 Physical Layer Attacks Attacks on the physical layer can be done by two ways 55 a Jamming Jamming is the process to interfere sensor frequency by transmitting same frequency in sensor operational area By doing this sensor node suffers deep interference and data are jammed b Tampering Tampering is the intentional way to damage and replace the existing sensor node Physical access to sensor node and electronic tampering may reveal sensitive information like encryption keys Jamming can be prevented by using spread spectrum technique or code spreading techniques Tampering can be prevented by using special packaging for the nodes These packaging are tampering free and very costly 7 10 3 2 Data link layer Attacks The possible attacks in data link layer are collisions and unfairness a Collisions Collision is process to collide
50. ltipath fading problems Terrestrial radio communications have lower signal degradation as compared to underwater radio communications b Commercial water quality sensors are expensive than other general terrestrial sensors c In salty water radio signals degrade significantly Hence there is always a problem in designing WSN d Underwater sensors have fewer lifetimes because of corrosion Maintenance and replacement of faulty parts are difficult 36 6 Lab Work Lab work was carried out in two phases In the first phase star network topology was created with gateway and two sensor nodes The maximum distance of wireless transmission between sensor node and gateway was determined and temperature measurement from both sensor nodes was monitored In the second phase distance of transmission was increased by configuring one sensor node as an intermediate router node and temperature was measured This 1s the case of multi hop network topology Lab work in the second phase was based on the Design2 described in the Chapter 3 1 2 Two numbers of NI WSN 3202 were taken as sensor nodes and one NI WSN 9791 was taken as a standard gateway Four standard AA size batteries were used to power one sensor node NI PS 15 was used to power another sensor node and gateway These WSN devices work on Zig bee standard 6 1 Phase l Star Network Topology Full configuration of WSN in star network topology is given in Figure 6 1 Both sensor nodes were di
51. multi hop network Measuring sensor node transfers data to the intermediate sensor node which is responsible to update and boost the signal and transfer it to nearby gateway For this design Zig bee IEEE 802 15 4 protocol and its features are used The advantage of using Zig bee is its high coverage range than that of W Fi As a result cost effective network using minimum devices can be designed 3 1 2 1 Technical Specification of Device Used in Design Proposed devices for design using Zig bee are from National Instruments NI WSN 3202 is used as measurement node and router node This device supports 4 analog input channels with 16 bit resolution and maximum sampling rate of I sample per second per channel It requires 30V DC power supply when configured as router and standard battery when configured as measurement node NI WSN 3202 1s also facilitated with 4 digital input output channels 3 NI WSN 9791 gateway 1s used as a sink node where we can connect our measurement station It uses 9 to 30V DC external power supply Communication range is 300 meters in USA and 150 meters in Europe It can support 8 end nodes in star topology and 36 end nodes in Mesh topology 14 Figure 3 4 shows NI devices used for design 2 NI WSN 3202 NI WSN 9791 Figure 3 4 NI devices used for design 2 3 1 2 2 Proposed Network Architecture When a single WSN 9791 is configured 1t can communicate with sensor nodes configured as measurement nodes or router n
52. nal strength reached to 60 sensor node was kept stationary in that place The distance where a sensor node was kept stationary was measured to be 23 1 meters Link quality fluctuation was measured up to 1024 samples and found that most of the time link quality ranges from 55 to 60 Thus maximum indoor distance of transmission was determined to be 23 1 meters with the link quality from 55 to 60 Figure 6 4 shows LabVIEW plot for link quality measurement Up to 168 samples sensor node was mobile then it was fixed in particular location and other samples were measured 0 ip Nun A daa T E Link Qualr l 1023 Time Figure 6 4 Time vs Link quality graph to determine maximum distance of transmission Sensor node powered with Alternative Current AC was kept stationary at 7 9 meters away from the gateway Figure 6 5 shows the top view location of both sensor nodes installed and their distance from gateway Figure 6 5 Top view of College to show where the sensor nodes and gateway were installed 40 6 1 3 Temperature Sensor Setup Two PT 100 temperature sensors were connected to first analog I O a10 of each sensor nodes and temperature fluctuations were measured through LabVIEW program Transmitter 2500 resistor 24 VDC power supply and PT 100 element were combined together to form a complete set of temperature sensor This set was then connected to sensor node as in Figure 6 6 PT 100 el
53. nd processed on CSS Non intrusive methods are used to process information in CSS and energy usage evaluation motor health conditioning and monitoring can be made for each motor in the plant Terminal quantities transferred to CSS from MCC s are generally current Is and Voltage Vs Is Vs speed estimator and resistor estimator are used to estimate corresponding speed Ss and resistance Rs For energy uses evaluation Vs Is Ss Rs and Name Plate Info NPI are needed NPI has information of motor types their operation range and rating By studying Machine Condition Prediction and Energy Plant Usage Evaluation plant manager can make a decision for replacing faulty and oversized motors and can work on the efficiency 29 improvement of under load motors Figure 5 1 gives the overall architecture of WSN based system architecture for closed loop industrial plant energy evaluation and planning system Enegry Usage Is Ss Evaluation gt Speed Estimator Vs Is Ss Rs NPI Vs g Stator Rs gt Resistance Motor Condition Estimator Monitoring Vs Is Rs NPI gt optional block Motor Nameplate Info NEI Central Supervisory station WSN Link gt Figure 5 1 WSN based architecture for closed loop industrial plant energy evaluation and Industrial Plant gt gt gt Plant Manager planning system 26 5 2 Case Study
54. networking management of and their security 6 Delivery of written thesis following the guidelines from TUC Task background Wireless networking has already penetrated the industries in various forms for data transfer communication and sensor networking With the increased interest for wireless networking in general in the industry naturally there is a need for looking into strategies of enhancing performance of wireless networking with respect to the possibility of integrating wireless networking with existing wired networks extending coverage distance of wireless networks etc Industrial users are interested predominantly in the following areas of study in wireless networking in general standards and their convergence to a form accepted by major actors in Adress Kjolnes ring 56 NO 3918 Porsgrunn Norway Phone 35 57 50 00 Fax 35 55 75 47 66 MIRATE the industries implementation of wireless networking management of wireless networks and their security Student category For SCE students with communication background and with good knowledge of National Instruments Wireless Sensor Networking Modules and their practical applications using LabVIEW Practical arrangements EIK HiT has the latest NI Wireless Sensor Networking Modules and has close interaction with the National Instruments Norway Necessary hardware and software will be provided by HiT Work will be performed in Sensor Lab and Flow Lab Possible interaction
55. o work then monitoring station may get a wrong picture of the situation 5 3 Case Study Ill This case study is based on the industrial filed visit in Yara nitric acid plant Porsgrunn The main purpose of the field visit is to understand how Wireless HART is integrated with DELTA V in order to monitor and control the process A team of an Engineer from Emersion Yara and TUC were involved for the coupling of devices 5 3 1 Background Wireless HART is a global International Electro technical Commission IEC standard 62591 for wireless communication in process industries 23 It is used for real time industrial process measurement and control applications Using Wireless HART one can implement self organizing mesh topology WSN Gateway host system wireless adaptor wireless field devices and wireless repeater are used to develop complete WSN system for process measurement and management DELTA V is a digital system used for the easy way to control supervise and data acquisition system in process industries 32 DELTA V is givena specific name of host system in Wireless HART based WSN system DELTA V communicates through open standards OPC OPC NET 3 0 to integrate with other plant functions and applications for real time and historical data transfer 33 Traditional process management system used Profibus or Field bus to connect sensors and actuators to the DELTA V system Development and research on WSN enable the feature of
56. obile NI WSN 3202 Battery powered Ethernet cable Laptop Figure 6 11 Full configuration in a multi hop network topology In MAX sensor node with serial number 159884B was selected Update Firmware tab was clicked and then from the drop down menu Mesh Router was selected Sensor node took some time before it was configured to router node Figure 6 12 shows the screen shot of MAX screen when one sensor node is updated as a router mode Circle in the screen shot clearly indicates Mesh Router mode File Edit View Tools Help a E My System 1 Restart fed Save E Refresh 4 Add Node X gt G gt 9 Data Neighborhood a Devices and Interfaces Td Scales Wireless Channel 14 9 6 Software gt j M Drivers ESTE A a Remote Systems Node Type Serial Number ID Last Communication Time Battery State Link Quality Network Mode Firmware Version E NI WSN9791 0156363C E NIWSN 3202 1598843 1 01 01 190403 56 53 OK Fair End Node NI WSN 3202 12 060 E NI WSN 3202 1598648 2 01 01 1904 03 56 52 No Battery Fair Mesh Router Y wsnVoltageNode 1 0 0f0 Firmware Figure 6 12 MAX window showing Network Mode for both sensor nodes where one sensor node is updated as Mesh Router 6 2 1 Determining Operation of Mesh Router Mode In order to determine the operation of a mesh router node battery powered sensor node was kept close to router node placed at 23 1 meters away from gateway as shown in Figure 6 13 At first bot
57. obility topology coverage area life time sampling rate cost energy etc WSN can be called perfect only when all the above mentioned implementation issues are addressed Before designing any WSN pre study 1s required to understand its measurement type its area of coverage mobility requirement and budget Then only design factors like topology deployment lifetime sampling rate and transmission protocol are studied and investigated In some of the applications of WSN like underwater monitoring bridge structural monitoring etc high transmission range is required in order to establish a link between measuring point to monitoring point In most WSN Industrial Scientific and Medicine ISM band frequency 1s used One of the limitations of ISM band is that output power of antenna cannot be more than 20 dBm 4 that means coverage area 1s limited WSN using National Instrument Zig bee devices can provide the highest range of 300 meters in America and 150 meters in Europe 3 at the cost of less sample rate as compared to Wireless Fidelity Wi Fi devices Coverage area can be increased by using router node Router node in WSN is a special type of measurement node which acts as repeaters These nodes are kept between end node and 10 gateway such that end measurement node first communicates with router node and then gateway This is a case of multi hop network Multi hop network can be changed to mesh network using more router nodes M
58. odes Maximum communication distance is up to 150 meters Because of omnidirectional antenna used in such device radiation pattern will be a circle of 300 meters diameter approximately Figure 3 5 shows the radiation pattern of single WSN 9791 using omnidirectional antenna 22 i ED wLs 3202 p WLS 9791 Vv 2 Radio Link 300 Meters g Figure 3 5 Radiation pattern of single WSN 9791 using omnidirectional antenna Advantage of omnidirectional antenna and property of sensor nodes that can be configured as router nodes are used together to extend the coverage For extending distance up to 300 meters two sensor nodes are sufficient One sensor node is configured as router and other is configured as measurement node Overlapping distance of 5 meters between router node and gateway is maintained such that WSN 3202 router node has range of 295 meters and WSN 9791 has range of 145 meters WSN 9791 gateway has 145 meters of range because only one side of its coverage area can be used Monitoring station or computer is connected to gateway Figure 3 6 shows the overall connection diagram The overall distance of measurement is not more than 450 meters in LOS g Design 2 GN 295 Meters ME Zigbee Network for WSN Symbol Description ta WSN 3202 router gt WSN 3202 measurement node Wireless link Laptop computer ed WSN 9791 Gateway Figure 3 6 Connection diagram of
59. oil Moisture Estimation and Wetlands Monitoring University of Puerto Rico Puerto Rico Aug 2005 6 Bray J Sturman C F BLUETOOTH 1 1 Connect Without Cables Prentice Hall Inc Second Edition USA 2002 2 4 7 Hill J L System Architecture for Wireless Sensor Networks University of California USA 2007 8 Castino J G Algorithms and Protocols Enhancing Mobility Support for Wireless Sensor Networks Based on Bluetooth and Zigbee Malardalen University Sweden Sep 2006 9 Gutierrez J A Callaway E H Jr Barrett R L Jr Low Rate Wireless Personal Area Networks IEEE Standards Information Network IEEE Press Second Edition Jan 2007 10 Wi Fi Alliance Discover and Learn 2011 http www w1 fi org discover and learn php Feb 2011 11 Javvin WLAN Wireless LAN by IEEE 802 11 802 11a 802 11b 802 11g 802 1 In 2011 http www javvin com protocolWLAN html Feb 2011 12 Lewis F L Wireless Sensor Networks Smart Environments Technologies Protocols and Applications University of Texas USA 2004 13 Akyildiz I F Wang X Wang W Wireless mesh network a survey Computer Networks 2005 47 445 487 14 National Instruments User guide and specifications NI WSN 9791 Ethernet Gateway National Instruments Corporations USA Nov 2010 15 Huang J H Wang L C Chang C J Deployment of Access Point for Outdoor Wireless Local Area Networks National Chiao Tung University Taiwan 2003 16 National Instruments Operating
60. other nodes in its transmission range Typical example of mesh network topology is given in Figure 2 4 Sensor Node Radio Network Figure 2 4 Mesh Network Topology In order to illustrate redundancy feature in mesh network topology Figure 2 5 is considered It consists of eight sensor nodes each connected on multi hop structure Sensor Node Radio Network Figure 2 5 Mesh network using 8 sensor nodes If measurement taken by node 8 has to be transferred to node 1 it can take either of following paths path 8 5 2 1 or path 8 5 4 1 or path 8 5 4 3 1 or path 8 6 4 2 1 or path 8 6 4 1 or path 8 6 4 3 1 or path 8 6 3 1 For an instance assume that current path used to flow signal from node 8 to node I is path 8 5 4 3 1 all of sudden node 4 fails to work Then mesh network automatically switch path 8 6 3 1 or path 8 5 2 1 giving redundant features This is illustrated in Figure 2 6 from the figure it can see that node 4 is failed and its corresponding paths are broken Sensor Node 8 Radio Network Figure 2 6 Mesh network illustrating redundant feature after its node 4 fails 2 2 4 Tree Network Topology The hybrid combination of peer to peer network and star network topologies is a tree network topology Tree network topology consists of root node and central hub Central hub is responsible for communicating with sensor nodes connected to them and root node Is responsible for merg
61. ound to be 6 V Statistical scopes for end node consist of green red and white color These represent minimum value maximum value and arithmetic mean value respectively 50 7 WSN Implementaion Issues European Science Foundation ESF has organized a workshop in April 2004 in order to investigate research in WSN and its practical implications in Europe Academic researchers and representative from different European country were participated and concluded with important dimensions of the sensor network design 40 Some of these dimensions of WSN design are discussed below These dimensions are the issues that need to be addressed while implementing WSN 7 1 Deployment The deployment of nodes in a given environment can be done in two ways They are random deployment and fixed deployment If nodes are installed haphazardly throughout the physical environment then they are considered to be random deployment whereas 1f nodes are installed with proper calculation and budgeting in a predefined location then it is considered as fixed deployment Nature of deployment is either continuous more nodes are deployed at any time during the operation of network or fixed time 7 2 Mobility In WSN the individual sensor nodes are assumed to be static However use of WSN in moving equipment demands the mobility feature When a sensor changes its position during operation time particular mobile node is allowed to use another frequency transmitted
62. p Link Qual Statistics Sensor Node 1 Statistics_s1 Aith Maxim PAY Minim PY FP Si Hist gay 50 50 Max Value S1 Min Value SI Mean Value S1 2 23 3786 20 9836 21 7751 40 40 y g Time of Maximum S1 Time of Minimum S1 2 30 E 30 for _ TE MENE ek EE 5 11 836 20 S 20 a a 10 I 10 k 0 I I 1 I i 0 I I I FS I I I I i 0 200 400 600 800 1023 v 55523579 59 65 Time Temperature Sensor Node 2 Statistics 52 Arith Maxim YY Minim AN Histogram 52 Hist al Max Value S2 Min Value S2 Mean Value _S2 50 30 23 1751 18 0622 20 6301 Cc 25 o 40 u a Time of Maximum 52 Time of Minimum 52 5 30 E w 17 7 5 o 33 3 w y t i w E I 1 l 1 1 I 1 l I l 0 5 10 15 20 25 30 35 40 45 50 Temperaute Figure 6 10 Front panel consisting Statistics tab for both sensors 1024 samples of temperature were measured For the temperature variation window was opened and closed in a room where sensor node 2 AC powered sensor node was kept 44 Sensor node I battery powered sensor node was kept in a big hall where the probability of temperature fluctuations was very less From Figure 6 9 it can be seen that temperature measurement for sensor node 2 fluctuated from 18 C to 23 1 C for the samples 0 to 1023 After 100 samples measurement the window of room was opened as a result temperature decreased After 500 samples time window was closed and temperature started to increase again
63. rature Phase Shift Keying European Science Foundation Gaussian Frequency Shift Keying Global System for Mobile Institute of Electrical and Electronics Engineering Industrial Scientific and Medicine International Organization for Standarization Intelligent Transportaion System Line of Sight Measurement and Automation eXplorer Motor Control Centers Micro Electro Mechanical system Sensors National Instruments Name Plate Info Offset Quadrature Phase Shift Keying Open System Interconnection Potential of Hydrogen Remote Authentication Dial In User Service Radio Frequency Special Interest Group Temporal Key Integrity Protocol Voltage Alternative Current Voltage Direct Current Wireless Access Point Wireless Distribution System Wireless Fidelity WireLess Sensor Wi Fi Protected Access Wi Fi Protected Access 2 Wireless Sensor Network Letters and expressions C dBm hp Is Kbps Mbps MHz mW Rs Vs uG Degree Centigrade deciBels Milliwatt horse power Current Kilobits per second Megabits per second Mega Hertz milli Watt Resistance Voltage micro Gravity I Introduction A Radio Frequency RF network which consists of transceivers sensors machine controllers microcontrollers and user interface devices 1s called Wireless Sensor Network WSN Sensors are spatially distributed entities those cooperatively monitor physical or environmental conditions such as vibration pressure motion temperature etc
64. re work Following future work are suggested to elaborate the findings of this study a b c Existing wired system installed in any applications can be replaced by WSN for reliability and performance testing co existence problems can be studied by installing Zig bee based WSN near to any other WSN based in ISM bands Monitoring application can be elaborated to monitoring and control application by integrating WSN with DELTA V process management system Small Network Management Protocol SNMP can be implemented in order to access data from any remote computers 60 d LABVIEW program can be improved by adding data logging facility database facility and alarm handling facility e Mesh network topology can be established between sensor node and gateway by adding one more router node 61 10 References 1 Santi P Topology Control in Wireless Ad Hoc and Sensor Networks John Wiley amp Sons Ltd First Edition England 2005 9 10 2 National Instruments User guide and specifications NI WLS ENET 9163 National Instruments Corporations USA Feb 2010 3 National Instruments User guide and specifications NI WSN 3202 National Instruments Corporations USA Nov 2010 4 Loy M Karingattil R Willams L ISM Band and Short Range Device Regulatory Compliance Overview TEXAS INSTRUMENTS USA May 2005 5 Villegas M A E Tang S Y Qian Y Wireless Sensor Network Communication Architecture for Wide Area Large Scale S
65. rectly communicating with gateway which in turn can be monitored by using LabVIEW program in computer Straight Ethernet cable was used to connect gateway with Laptop NI WSN 9791 AC Powered Mobile NI WSN 3202 Battery powered Ethernet cable Fixed NI WSN 3202 AC Powered Figure 6 1 Full configuration in a star network topology 6 1 1 Configuration Steps Coupling of WSN devices and its monitoring through LabVIEW programming are given step wise below 1 PC containing LabVIEW program was taken 3 NI WSN software that came along with device package was installed NI WSN 9791 was connected to PC using straight Ethernet cable Measurement and Automation eXplorer MAX was clicked from the tools button of LabVIEW window Remote Systems in MAX was expanded and NI WSN 9791 was detected by MAX NI WSN 9791 gateway was selected WSN nodes tab was clicked and after that add node tab was clicked Type of sensor node 1 e NI WSN 3202 device was selected serial number of the device from device sticker was typed and in ID Number 1 was entered After this Finish button was clicked Another NI WSN 3202 was added by clicking Add Another button and step 7 was repeated again but in ID Number 2 was entered ID number ranges from I to 50 and each device were given with unique ID number Now nodes were connected automatically to the gateway Figure 6 2 is the screen shot of MAX screen when both sensor nod
66. rnnnnnnnnrrrnnrnnnnnnnnsnnne 46 6 22 LabVIEW Programming and Data Interpretation ssrrnrrnnnrnnnnnnnnnrnnnnnnnnnnnnrnnnrnnnnnnnnrrnnnnnnnnnnnnsnnne 48 7 WSN IMPLEMENTATON ISSUES se 51 Jl DOMENE 51 T2 MOB arte 51 dio COST SIZE AND EINER GY SPR 51 14 COMMUNICATION MODALITY PP 52 To MINTER AST O Eo 52 7 6 NETWORK TOPOLOGY AND COVERAGE gen 52 Fel NETWORK SIZE cios 52 A eo e o i a o no eea ana onina oaa 32 RH SAMENG RATE orae E E E seas needs 53 7 10 FER RAA 53 7 10 1 SECULU Y M ONG OAN IU AEREE OE E OET aa 53 7 10 2 ECU Mina mn nas a 54 7 10 3 Protocol Stacks Security Threats and Prevention rrrrrnrrrnnrnnnnnnnnnnnnnrnnnnnnnnnrnnrnnnnnnnnnrrnnnnnnnneere 55 8 DIS CU SST Nina ASAS A Aa SATA A DA 58 Bal TPROTOCOLAND TOPOLO6 Vd co aras 58 0 2 DESIGNAN e DEINE 58 8 3 TRANSMISSION DRENERE 59 SECURITY TRE 59 9 CONCLUSION AND FUTURE WORK vgs 60 VNR 60 DL FEN RR 60 0 REFERENCES Se 62 DE APPENDICES iieii i aaa aeaii aaa Eaa aiaiai 66 Overview of tables and figures Table 2 1 Blu tooth Specilications lucir 14 Table 2 22 EE TND 14 Tabl 2 3 Wi FL SpPECI CAOS AA A ab 15 Table 3 1 Cost estimation of WSN using NI Wi Fi Modules ooocccccccccnnncnoccccnnnnncnnnnnnos 21 Table 3 2 Cost estimation of WSN using NI Zig bee Modules oooocccccccccnnnnnnccnnnnnnnnnnnnnnos 24 Table 4 1 Advantages and Disadvantages of WSN ooooooooccccnnnnnnnonnnonccnnnnnnonononannnnncnnnnnonnnnos 27 Table4 2 Source tor mte
67. rom this study a b c d e 2 Zig bee provides high range of transmission but less data rate Wi Fi provides high data rate but less transmission distance Bluetooth range and data rate are moderate as compared to other two Mesh network topology support multi hop networking where a node can talk to adjacent nodes This feature allows redundancy in the system Mesh topology is used when WSN needs to be implemented in the large geographical region Wi Fi based WSN design is expensive as it demands the large number of AP For the same transmission distance Zig bee based design is cheap but suffers from less sampling rate problem Tradeoff between sampling rate and transmission distance can be solved by using National Instrument module CompactRIO along with Wi Fi DAQ and third party Zig bee transmission module 23 1 meters of transmission range between two WSN 3202 sensor nodes and WSN 9791 gateway in star topology was significantly increased to 47 1 meters using multi hop topology where one sensor node was configured as the router node Single node theoretical range of 150 meters was not achieved due to nature of an operating environment LOS cannot be maintained between a gateway and sensor nodes and signal suffers reflection refraction scattering multipath fading etc Authentication and Encryption are two key security components of wireless networks Each layer of ISO OSI suffers different types of attack 9 2 Futu
68. rterenca usanne A ER EEA 27 Table 6 1 Elements used in LabVIEW code and their functions cccccccnnnnnnnnnnnnnonnncnnnnnnnnnos 43 Fure jl DETS VISNE 12 Proure 2 27 Peer to Peer Network TOpOlO gy acia 15 Fore 2d Mr Netw AAA O A ang 16 Fisure 2 4 Mesh Network Topology siii td 16 Figure 2 5 Mesh network using 8 sensor NOES rrrrnnvnnnvvrrrrrrnrnnvnnnnnnrrrrrnnnnnnnnneeerssssssnnnnnnssnne 17 Figure 2 6 Mesh network illustrating redundant feature after its node 4 fails 17 Fore 1 ree NetWork TOP Y rara dra 18 Fisure JE NTdevices used Tor des On Lae 20 Figure 3 2 Radiation pattern of single WAP 3711 using omnidirectional antenna 20 Figure 3 3 Connection diagram of Wi Fi WSN using WDS system occcccccccnnonoccccnnnnncnonnnnos 21 Feed NIdeV Cees Used Od NL 22 Figure 3 5 Radiation pattern of single WSN 9791 using omnidirectional antenna 25 Figure 3 6 Connection diagram of Zig bee WSN using sensor node as router 23 Figure 5 1 WSN based architecture for closed loop industrial plant energy evaluation and Pans EN var 30 Figure 5 2 Sensor node layout on Golden Gate Bridge 31 oocccccccnnnnnoccnnnnnnnnnnnnnnno 31 Figure 5 3 Existing process management system and Wireless HART based process mA tl rta dador 33 Figure 5 4 Engineers busy on random installation of wireless HART in Yara Porsgrunn 33 Figure 5 5 Possible approach
69. s a Desingn 1 ON Wi Fi Network for WSN Symbol Description ph WAP 3701 3711 Laptop computer WLS 9163 Figure 3 3 Connection diagram of Wi Fi WSN using WDS system First WAP 3711 1s used to communicate with sensor node WLS 9163 and all other WAP 3711s are used as repeaters The overall distance of measurement is not more than 330 meters in Line of Sight LOS 3 1 13 Cost Estimation Device implemented in first design their individual cost and grand total after VAT and discount is given in the Table 3 1 Table 3 1 Cost estimation of WSN using NI Wi Fi Modules Cost Estimation for Design 1 Wi Fi Network using NI devices O md a NI WLS 9163 IEEE 802 11b g Carrier for C 3699 00 2 7398 00 Series Modules NI WLS 9234 12999 00 25998 00 NI PS 15 Power Supply 24 VDC 5 A 100 1699 00 7 11893 00 120 200 240 VAC Input NI WAP 3711 3701 5999 00 48993 0 23820 00 Total cost for Wi Fi based design is approximately 119102 00 NOK This cost has been calculated with the reference to price list provided by National Instruments which is given in Appendix 2 21 3 1 2 Design 2 Using Zig bee In this design concept of intermediating AP used as repeater is avoided instead intermediate sensor node is configured as a router One more sensor node is placed in between the existing sensor node and gateway Such that overall coverage will be sum of both This design is based on the concept of
70. smission range of 300 meters both designs can be used However Design I seems more expensive The total approximated cost for Design 1 implementation is around 119102 00 NOK whereas the cost for implementing Design 2 is very less and is around 18131 00 NOK Design I uses Wi Fi DAQ which provides high bandwidth and sampling rate of 51 2 K samples per second per channel Nevertheless Design 2 uses Zig bee DAQ which provides very fewer sampling rates of I sample per second per channel For any physical quantity demanding high sampling rate Design 2 fails and for any measurement that requires transmission distance more than 330 meters Design 1 fails This suggests that implementation of WSN in the area that demands both range and sampling rate is practically difficult to design especially when ISM band is considered and vibration is the interest of measurement The tradeoff between a range and sampling rate can be solved by modifying the design Use of National Instruments CompactRIO device along with third party modules 48 can give both high transmission range and high sampling rate Doing this we can have a high sampling rate of Wi Fi Data Acquisition DAQ as Design 1 high transmission distance of Zig bee as in Design 2 and new design will be cost effective For this NI WLS 9234 can be used as DAQ device and SEA cRIO Zig bee 49 can be used as transmission modules 58 8 3 Transmission Distance In star network topology the maxim
71. ssues Lab work was carried out to understand how intermediate sensor node configured as router node can help to extend the WSN coverage I would like to express my deep sense of gratitude to my supervisors Prof Saba Mylvaganam and Assoc Prof Hans Petter Halvorsen I would also like to acknowledge invaluable support from National Instrument employees Frode Skulbru and Tom Arne Danielsen Special thank goes to employee from Yara nitric acid plant Porsgrunn for organizing effective field visit of their WSN systems I would like to thank Telemark University College Porsgrunn Norway and National Instrument Norway for giving me opportunity to work in this very interesting and relevant topic It has been a great experience working on practical based project which has taken me to the new level of understanding the concept of WSN Finally I would like to thank everybody who has supported me directly and indirectly Telemark University College Porsgrunn June 3 2011 Rabin Bilas Pant Nomenclature Abbreviations AC AES AP BAN BPSK CSS DAQ DO DPSK DOPSK ESF GFSK GSM IEEE ISM ISO ITS LOS MAX MCC MEMS NI NPI OQPSK OSI pH RADIUS RF SIG TKIP VAC VDC WAP WDS Wi Fi WLS WPA WPA2 WSN Alternative Current Advanced Encryption System Access Point Boday Area Network Binary Phase Shift Keying Central Supervisory Stations Data Acquisition Dissolved Oxygen Differential Phase Shift Keying Differential Quad
72. ta from farthest sensor nodes to the monitoring station Vibration measurement was taken at I KHz rate 29 Two types of accelerometer sensors were used for collecting the vibration data from a bridge These were ADXL 202E and SD 1221L Each type of accelerometer sensors was responsible for the different range of vibration detection Mica2 motes were used as sensor nodes for storing and communicating data Mica2 consists of ATmelATmega 128L microprocessor 128 KB of programmable memory 4KB of RAM Chipcon CC1000 radio chips and operates at 8 MHz with data rate of 28 4 Kbaud 30 Sensor layout on the Golden Gate Bridge 1s shown in Figure 5 2 Sausalito north EEE J ee me y O D Ae O aa eee dl A Pa 4200 fi 56 nodes 8 nodes Figure 5 2 Sensor node layout on Golden Gate Bridge 31 5 2 3 Data Acquisition Issues Some data acquisition issues 30 related to structural health monitoring are given below a For every sensor reading static noise is associated with it Less static noise floor gives high accuracy measurement Structural health monitoring application required the detection of signal down to 500uG for a distortion less measurement 31 b Structural health monitoring applications require high frequency sampling in order to avoid the variations in sampling interval jitter c In order to monitor a large area multi hop network is established In such cases 1f any node fails t
73. teway were installed and new path followed by end node after resettidg ococoooooooncnncnnnnnononononnnnnnnnnonononannnnnnnnnnnos 4 Figure 6 14 Time vs Link quality graph to observe improvement in link quality after end DOE NS Den Tesla a S N 47 Figure 6 15 MAX window showing Excellent link quality after the end node was reset 48 Figure 6 16 Top view of College to show where the sensor nodes and gateway were installed and Mer separa on NE je 48 Figure 6 17 Block diagram of LabVIEW program used to measure temperature from end NM NN 49 Figure 6 18 Front panel consisting temperature statistics for end node and link quality statistics for both end node and router node rrrrrrrrnnnrrrrrrvvvrrnnnnnerrrrrnnnvrnnnnnssssssrnnsrnnnnnesssssseene 49 Figure 7 1 WSN Architecture with security components 43 ccccccccnnnnnncccnnnnnnnnonnnno 54 Figure 7 2 Sensor node protocol stack 46 oooooooonnnccnnononononnnoccnnnnnnnonononncncccnnnononnnnos 55 Preface This thesis is prepared as an obligatory requirement to be graduated as master in System and Control Engineering at the Telemark University College The entire work was carried out in Sensor Lab Flow Lab and Process Hall The necessary technical information and equipment were provided by technical department of College The main focus was given to understand the current standard scenario of WSN its advantages and disadvantages and security i
74. to avoid flooding client puzzle technique can be used In this technique client must solve the puzzle send by the server each time they demand connection De synchronization can be avoided by a proper authentication scheme 57 8 Discussion This chapter discusses the result obtained in this study The discussion is mainly focused on protocol and topology comparison between design I and design 2 transmission distance and security issues 8 1 Protocol and lopology Zig bee seems to be the best protocol if range and power are considered while Wi Fi seems to be the best protocol if data rate is considered Zig bee supports range up to 150 meters at the cost of lower data rate of 250 Kbps whereas Wi Fi supports data rate up to 54 Mbps at the cost of lower range of 30 meters However newest W1 Fi technology can support 125 meters range with data rate of 200 Mbps but this technology is not widely implemented Bluetooth has moderate data rate of 3 Mbps with a maximum range of 100 meters Mesh network topology has redundant feature Mesh topology 1s the combination of numbers of multi hop networks where any node can talk with other nodes in the network Thus it is used in such WSN where nodes are to be distributed in the large geographical area Another network topology like star tree and point to point cannot give the redundant feature and their coverage area are limited to the range of AP 8 2 Design 1 vs Design 2 For any tran
75. two data packets it is somehow similar to jamming In collision whole packets can be damaged b Unfairness MAC priority scheme is harmed or ill treated such that sensor node cannot maintain the real time deadlines causing service degradation Collision can be prevented by using collision detection technique or error correcting codes Unfairness is prevented by using small frames such that channels are captured only for small time interval and released very soon 7 10 3 3 Network layer Attacks Attacks in the network layer give some serious misinformation which results in packet dropping wrong routing information and misdirection All the possible attacks in the network layer can be avoided by following methods a Use of link layer encryption and authentication b Use of redundant network to form a multi path routing c Use of unique key in order to establish communication between a nodes and base stations 7 10 3 4 Transport layer attacks Flooding and de synchronization are the terms used to describe the attacks in the transport layer a Flooding By flooding attacks attackers send many connection establishment requests System responds to each request by allocating memory As the result memory resources of the system are exhausted b De synchronization In this type of attacks control flags and sequence numbers are modified this may lead infinite cycle of retransmitting previous erroneous message 56 In order
76. uetooth in terms of raw data rate version and modulations scheme is included in Table 2 1 Bluetooth has low latency high throughput operation with minimum channel hopping period of 600 microseconds 7 Since it has very less channel hopping period all nodes should be synchronized very fast this adds complication in some WSNs where nodes sends data very slowly Bluetooth based WSN takes 2 4 seconds for connection establishment that is its disadvantages Table 2 1 Bluetooth Specifications 8 GFSK 74 Modulation GFSK GFSK GFSK DOPSK SDPSK I Mbps I Mbps I Mbps 2 1 2 Zig bee Zig bee is based on IEEE 802 15 4 standard which provides ultra low complexity low cost and extremely low power wireless connectivity for inexpensive and portable device 9 Zig bee uses 868 MHz in Europe 915 MHz in USA and Australia and 2 4 MHz in other parts of world The data rate ranges from 20 Kbps to 250 Kbps modulation scheme is either Binary Phase Shift Keying BPSK or Offset Quadrature Phase Shift Keying OQPSK Zig bee consumes less power than Bluetooth and Wi Fi Technical specifications of Zig bee are given in Table 2 2 Table 2 2 Zig bee specification Symbol Rate Modulation Ksymbol S Frequency Bit Rate 868 868 6 2400 2835 2 1 3 Wi Fi Wi Fi is based on IEEE 802 11 protocols which works on ISM bands of 2 4 GHz and 5 MHz 10 These are more secured and high bandwidth transmission techniques The bandwidth and range o
77. um distance of communication between sensor node and gateway was found to be 23 1 meters with link quality not less than 55 Theoretical transmission distance is 150 meters at LOS Experiment was set up inside the hall where LOS between sensor nodes and gateway was not possible to maintain Instruments machineries placed in the hall and closed surface of room where the gateway was placed creates diffraction reflection refraction scattering shadowing and multipath fading of a signal such that their link quality degraded and overall transmission distance was decreased to 23 1 meters rather than 150 meters In order to increase the transmission distance at least two sensor nodes are needed where one sensor node act as a router and other node act as an end node Router node that acts as the repeater links the end node and gateway In this mode the link quality of the end node is not depend on the position of gateway 1t entirely depends on position of the router node Hence end node can be moved away from router node until the link quality between the end node and router node becomes 55 The overall transmission distance was found to be sum of distance from end node to router node and distance from router node to gateway The total transmission distance was increased from 23 1 meters to 47 1 meters at fair link quality of 55 Thus coverage extension was successfully achieved using a router node concept but range suggested by design 2 was not
78. ur case studies for different application of WSN will be presented and discussed in chapter five Chapter six will present detail of lab work carried out in two phases The phase I will deal with star network topology and the phase II will deals with coverage extension using router mode sensor node Chapter seven will discuss some WSN implementation issues where security issues will be discussed more deeply Discussion of lab work and major associated theories will be discussed in chapter eight Chapter nine will point out the conclusion of a thesis along with its perspectives to elaborate it in the future 2 Standard Scenario of WSN The diversity in application areas and the specific properties of WSN demands some sort of regularization from sensing point to end users These can be defined in terms of network topology bandwidth offered transmission protocols and sensor s compatibility Following article deals with the current standard scenario of WSN and its essential technical details for transmission protocols and network topology Figure 2 1 is considered for a detailed structure of WSN This figure 1s distinctly categorized into two sections Classical Infrastructure and Sensor Networks Classical Infrastructure Sensors Networks ome cp Base station Acess Point Internet Cloud D User Interface S Network Switch Sensor Node Z HighBadwidth Radio Link Low Bandwidth Radio Link Ethernet Link
79. using wireless technology to replace cabling installation such that relocation and installation have become very easy 5 3 2 Existing Process Management System YARA has installed DELTA V for process management All the sensors were connected to DELTA V process management system through a wired system like an optical fiber field bus etc In order to avoid cable and make the process management system more robust non critical process data monitoring was planned to be replaced by wireless link They have chosen Wireless HART system based WSN Integrating Wireless HART allows devices to communicate wirelessly with existing DELTA V system Self organizing mesh network was created and data from all sensors were transferred to DELTA V system Figure5 3 gives an overview of existing process management system and Wireless HART based process management system 32 Existing Process E Wireless HART based Management System Process Management System Figure 5 3 Existing process management system and Wireless HART based process management system 34 5 3 3 Coupling Engineers first tried to install Wireless HART sensors randomly The gateway was placed on the room which was located on the 4th floor of the plant building and sensor nodes were placed on the roof of 7 floor where pumps and other machineries were installed as shown in Figure 5 4 The rough distance between sensor nodes and gateway was about 60 meters One sensor node with vibr

Wireless Sensor Networking with Lab

Contents

Download Pdf Manuals

Related Search

Related Contents