Iraqi Rivers Pollution Monitoring System Based on Underwater Wireless Sensor Networks

Ministry of Higher Education and

Scientific Research

University of Technology

Computer Engineering Department

Iraqi Rivers Pollution Monitoring

System Based on Underwater Wireless

Sensor Networks

A Thesis

Submitted to the Computer Engineering Department University of Technology in Partial Fulfillment of the Requirements for the Degree of Master of Science in

"Computer Engineering"

By

Ahmed Mosa Dinar

(B.Sc. 2007)

Supervised by

Assist.Prof. Dr Mohammed Najm Abdallah

Nov. 2013 Muh. 1434

يرفع الل�ه ال�ذين آمنوا منكم ﴿

﴾وال�ذين أوتوا العلم درجات

( 11: المجادلة )

DEDICATION

This Thesis is dedicated to:

My Mother

and every mother…

My Father

and every father…

Ahmed Mosa

November / 2013

ACKNOWLEDGEMENTS

All praises and thanks are due to Almighty "Allah" who

enabled me to complete this task successfully.

My appreciation goes first to my supervisor Assist. Prof. Dr.

Mohammed Najm Abdallah for giving me advices and guidance

for the completion of this thesis.

I would like to thank the staff of the Computer

Engineering Department for the great help they have introduced

to me.

Thanks for the Environmental Research Center at the

University of Technology and the Environmental Baghdad

Department at ministry of environment for helping me and

giving me useful information.

Last, I would like to thank my family and friends for their

support during my study.

Ahmed Mosa

November / 2013

Supervisors Certification

I certify that this thesis entitled (Pollution Monitoring System Based on

Underwater Wireless Sensor Networks) was prepared by (Ahmed Mosa

Dinar) under my supervision at the Computer Engineering Department /

University of Technology in partial fulfillment of the requirements for the

degree of Master of Science in " Computer Engineering" .

Signature:

Name: Dr. Mohammed Najm Abdallah

Title: Assistant Professor

Position: Supervisor

Date: / 9 / 2013

In view of the available recommendation, I forward this thesis for debate by the examination committee.

Signature:

Name: Prof.DR. Salih M. Al-Qarraawi

Title: Head of Department

Date: / 9 / 2013

V

Linguistic Certification

This to certify that this thesis entitled (Pollution Monitoring System

Based on Underwater Wireless Sensor Networks) was prepared by

(Ahmed Mosa Dinar) under my linguistic supervision.

Its language was a mended to meet the style of English language.

Prof. Dr. Arkan KH. Husain Al -Taie

Linguistic Supervisor

22/9/2013

Certificate of the Examination Committee

We certify, as an examination committee, that we have read the thesis

entitled " Iraqi Rivers Pollution Monitoring System Based on

Underwater Wireless Sensor Networks" , and examined the student

" Ahmed Mosa Dinar" and found that the thesis meets the standard for

the degree of Master of Science in Computer Engineering.

Approval of the College of Computer Engineering

Signature: Title : Assist. Prof. Name : Dr. Mohammed Najim Abdallah

(Supervisor) Date : / /2013

Signature: Title : Assist. Prof. Name : Dr. Mohammed Yousif Hassan (Member) Date : / /2013

Signature: Title : Professor Name : Dr. Sideeq Yousif Ameen

(Chairman) Date : / /2013

Signature: Title : Professor Name : Dr. Salih M. Al-Qarraawi (Dean) Date : / /2013

Signature: Title : Lecturer Name : Dr. Muayad Sadik Croock

(Member) Date : / /2013

I

Abstract

The pollution monitoring system describes the processes and

activities that need to implement through monitor the quality of specific

environment. In this thesis those activities and processes are presented by

assuming two important areas, the design of underwater wireless sensor

network and a proposal water quality index in Iraq using fuzzy inference

system.

In the first area (design of underwater wireless sensor network),thesis

focus on the performance assessment of five underwater wireless sensor

network (UWSN) Media Access Control (MAC) layer protocols namely;

(Broadcast MAC, Aloha, R-MAC, FAMA, and UWAN_MAC). These

protocols are used for the water environment in terms of energy

consumption and received throughput. Three of them are nominated

based on some previous assessment. They were compared in terms of

total drop packets and average end to end delay. Subsequently, an

investigation of the facts about what is the robust and energy efficient

routing protocol in underwater wireless sensor networks is presented.

This is done by comparing three important routing protocols, namely;

Vector-Based Forwarding Protocol (VBF), hop-by-hop vector-based

forwarding (HHVBF) and Vector-Based Void Avoidance (VBVA). The

performance evaluation of these protocols is achieved using metrics

energy consumption, average end-to-end delay and Packet delivery ratio.

Results of these evaluations and comparisons prove that UW-MAC

protocol is the most suited one with geographic routing protocols

especially VBF. Assessment is carried out by using Aqua-Sim simulators

for underwater sensor networks and NS2 based simulator installed in

Linux environment.

II

A robust and flexible index was made to be used as baseline for

all pollutants that need to be monitored. This has been done with fuzzy

inference system to present a work aimed to propose water quality index

in Iraq. Fuzzy logic is used for developing the traditional environmental

indices to eliminate the routine in assessment of water quality. As well as

to overcome the largest problems uncertainty and subjectivity in

traditional way; we called "FWQI". To evaluate the performance of the

index suggested with the real conditions, the case study was conducted on

Tigris River, Iraq. The data of water quality from different sampling

positions were taken. It was found that the achieved results from Fuzzy

Water Quality Index (FWQI) are less than the results obtained from the

regular method in final score of pollution by approximately 41.64% for a

specified period.

III

Table of Contents

Abstract ……………...………………………………....................... I

Chapter One: Introduction

1.1 Introduction 1

1.2 1.2.1 1.2.2 1.2.2.1 1.2.2.2 1.2.2.3

Research Motivation and Challenges Motivation Research challenges UWSN challenges Network Implementation challenges Traditional water quality index challenges

3 3 4 4 6 7

1.3 Literature Review 7

1.4 Aim of work 12

1.5 Thesis Layout 13

Chapter Two: Theory of Pollution Monitoring System

2.1 Introduction 14

2.2 2.2.1 2.2.2

Wireless Underwater Sensor Networks Terrestrial Sensor Networks versus Underwater Sensor Networks UWSN Applications

14 14

15

2.3 Comparison of RF, Optical and Acoustic Communication Underwater

17

2.4 Underwater Sensor Network Components 18

2.5 2.5.1 2.5.2

Communication Architecture Two-dimensional Underwater Sensor Networks Three-dimensional Underwater Sensor Networks

21 21 23

2.6 Protocol stack for underwater acoustic channels 24

IV

2.6.1 2.6.1.1 2.6.1.2 2.6.1.3 2.6.1.4 2.6.1.5 2.6.2 2.6.2.1 2.6.2.2 2.6.2.3

MAC Layer Protocols Floor Acquisition Multiple Access (FAMA) Routing enhanced MAC (R-MAC) Underwater Acoustic MAC (UW-MAC) Aloha Broadcast MAC Routing protocols Vector based forward Hop-by-Hop Vector-Based Forwarding (HHVBF) Vector-Based Void Avoidance (VBVA)

26 27 29 31 34 34 34 35 36 39

2.7 2.7.1 2.7.2 2.7.2.1 2.7.2.2 2.7.2.3 2.7.2.4

Water quality assessment Background Implementing Materials and methods Fuzzy inference system Membership functions Fuzzy set operations Inference rules

42 42 44 44 45 45 46

Chapter Three: UWSNs Simulation and Network Result

3.1 Introduction 47

3.2 3.2.1

UWSN simulation and Simulation Design Fidelity and flexibility testing of Aqua-sim

47 52

3.3 3.3.1 3.3.1.1 3.3.2 3.3.2.1 3.3.3 3.3.3.1

Performance Assessment of UWSN Protocols UWSN MAC Layer Protocols Assessments and Simulation Results UWSN Routing Protocols Assessments and Simulation Results Final Design Assessment and Simulation Results

54 54 54

59 59

63 63

3.4 Summary of contributions and achievements 69

V

Chapter Four: Development of a water quality Assessment Using FIS

4.1 Introduction

71

4.2 Calculation of Traditional WQI in Iraq 71

4.3

Fuzzy inference systems, step by step 73

4.4 Development of the Fuzzy Water Quality Index (FWQI)

78

4.5 Results and Discussion of FWQI 79

4.6 Validity of FWQI 83

Chapter Five: Conclusion and Suggestions for Future Work

5.1 Conclusion 85

5.2 Suggestions for Future Work 87

References 88

Appendices

Appendix A 99

Appendix B 113

XI

List of Figures

Chapter Two Pages No.

Figure (2.1): Internal organization of an underwater sensor node.

Figure (2.2): Architecture for 2D underwater sensor networks.

Figure (2.3): Architecture for 3D underwater sensor networks.

Figure (2.4): The sensor network protocol stack.

Figure (2.5): Hidden terminal example.

Figure (2.6): The operation of the UWAN-MAC mechanism.

Figure (2.7): The transmission packet structure and the listen duration for the UWAN-MAC mechanism. Figure (2.8): VBF using single routing pipe. Figure (2.9): HH-VBF with per-hop vector computing.

Figure (2.10): An example of void node.

Figure (2.11): Two mechanisms of VBVA.

Figure (2.12): Input–output map for the river water quality problem in a fuzzy inference system.

20

22

23

25

30

33

34

37

39

40

41

44

XII

Chapter Three Pages No.

Figure (3.1): Main Stations of a Pollution Monitoring System

Figure (3.2): Evolution of Monitoring Systems.

Figure (3.3): Relationship between Aqua-Sim and other packages of NS-2

Figure (3.4): Class diagram of Aqua-Sim.

Figure (3.5): Network topology for fidelity testing.

Figure (3.6): Throughput with fixe input traffic per node.

Figure (3.7): Throughput with fixed total input traffic.

Figure (3.8): Simulation Scenario of MAC Protocol Test.

Figure (3.9): Energy Consumption of UWSN MAC Protocols.

Figure (3.10): Received Throughput of UWSN MAC Protocols.

Figure (3.11): Average End To End Delay of UWSN MAC Protocols.

Figure (3.12): Total Drop Packets of UWSN MAC Protocols.

Figure (3.13): Packet Delivery Ratio of UWSN Routing Protocols.

Figure (3.14): Average End-To-End Delay of UWSN Routing Protocols.

Figure (3.15): Energy Consumption of UWSN Routing Protocols.

Figure (3.16): Network Topology of Final Design.

Figure (3.17): Total Drop Packets of UWSN in Final Design.

Figure (3.18): Energy Consumption of UWSN in Final Design.

Figure (3.19): Average End to End Delay of UWSN in Final Design.

Figure (3.20): Received Throughput of UWSN in Final Design.

48

48

50

51

52

53

53

56

57

57

58

58

61

62

62

65

66

67

67

68

Chapter Four Pages No.

Figure (4.1): Membership functions for PH, DO and FWQI parameters.

Figure (4.2): Fuzzy language Rules for the inputs PH and DO.

Figure (4.3): Fuzzy inference diagram for the water quality scoring problem

with two variables and five rules.

Figure (4.4): A surface graph representing the interactions between DO, PH

and the final index value.

74

75

77

78

XIII

Figure (4.5): Results for water quality indexes for Tigris River within Baghdad city from 2004-2010 Figure (4.6): Comparative between indexes in June 2007.

Figure (4.7): Comparative between indexes in August 2007.

Figure (4.8): Comparative between indexes in January 2008.

81

82

82

83

XIV

List of Tables

Chapter One Pages No.

Table (1.1): Questionnaire in some pollution monitoring centers 5

Chapter Two Pages No.

Table (2.1): Differences between terrestrial and underwater sensor networks. Table (2.2): Comparison or RF, optical and acoustic communication underwater.

15

19

Chapter Three

Table (3.1): Simulation Parameters of MAC protocols Test.

Table (3.3): Simulation Parameters of Routing Protocols Test.

Table (3.5): Final design Simulation Parameters

Table (3.7): Available bandwidth for different ranges in UW-A

channels.

55

60

64

69

Chapter Four

Table (4.1): Computed WQI values for Tigris River within Baghdad

city from 2004-2010.

Table (4.2): Parameters for membership functions used in the fuzzy

inference system

73

80

XV

List of Abbreviations

Acronym Definition

ACK Acknowledgment

ACK -REV Acknowledgment -Reservation

ANOVA Analysis Of Variance

Aqua-Sim Aquatic Simulator

AUV Autonomous Underwater Vehicles

CDMA Code Division Multiple Access

CMU Carnegie Mellon University

CSMA Carrier Sense Multiple Access

CTS Clear To Send

EM Electromagnetic

FAMA Floor Acquisition Multiple Access

FDMA Frequency Division Multiple Access

FIS Fuzzy Inference Systems

FSO Free-Space Optical

FWQI Fuzzy Water Quality Index

GPRS General Packet Radio Service

GPRS General Packet Radio Service

GSM Global System for Mobile

GUI Graphical User Interface

HH-VBF Hop-by-Hop Vector-Based Forwarding

MAC Media Access Control

MACA Multiple Access with Collision Avoidance

XVI

MACAW Multiple Access with Collision Avoidance for Wireless

MEMS Micro Electro-Mechanical Systems

NS-2 Network Simulator version number 2

OTcl Object oriented extension of Tcl

pH Potential of Hydrogen

QoS Quality of Service

RF Radio Frequency

R-MAC Routing enhanced MAC

RTS Request-To-Send

SDV Small Delivery Vehicle

SYNC Synchronization

Tcl Tool Command Language

TDMA Time Division Multiple Access

UW Under Water

UWAN Underwater Acoustic MAC

UWSN Underwater Wireless Sensor Networks

VBF Vector Based Forward

VBVA Vector-Based Void Avoidance

WQI Water Quality Index

WSN Wireless Sensor Networks

Chapter One

Introduction

Chapter One……………………………………….…………………….Introduction

1

Chapter One

Introduction

1.1 Introduction

Sensing is a technique used to collect information about a physical object or

process, including the happening of events such as changes in state like dropping in

temperature or pressure. An object performing a sensing task is called a sensor. For

example, the human body is supplied with sensors that can capture optical

information from the environment (eyes), acoustic information such as sounds (ears),

and smells (nose). These are examples of remote sensors, i.e. they do not require

touching the monitored object to gather information. From a technical view, a sensor

is a device that translates events or parameters in the real world into signals that can

be determined and analyzed [1].

With the new developments in micro electro-mechanical systems (MEMS)

technology, in addition to wireless communications, and digital electronics, the

design and development of cheap, low-power, multifunctional sensor; tiny nodes that

communicate untethered in short distances are applied. The continuously increasing

capabilities of these small sensor nodes, including sensing, data processing, and

communicating, enable the realization of wireless sensor networks (WSNs) based on

the effort of a large number of sensor nodes [2].

So, a WSN is a self-configured network that is composed of a large number of

small sensor nodes. The fields of WSNs influence the world with their potential to

enhance human lives. A WSN is able to sense information, processing, and transmit

information to other nodes through proper communication. Sensor nodes were

developed for deployment on land, and there are no native nodes for underwater

environment until now. Terrestrial nodes enclosed in specially designed containers

deploy the nodes in water but they do not exploit underwater circumstances. These

waterproof nodes are susceptible to issues such as localization, communication


2

overhead (i.e. electromagnetic waves are vulnerable over long distances in

underwater environment), bandwidth, dynamic topology, and node flexibility.

Henceforth, using nodes bounded in containers is not a perfect solution to monitor

the underwater environment and is not recommended for placement in underwater

environments [3].

Lately, wireless sensor networks were suggested for placement in underwater

environments where a lot of applications like aquiculture, pollution monitoring,

offshore investigation, etc. would advantage from this technology.

Design and construction of UWSNs as a communication network is extremely

difficult. As a value, that is valid for earthly WSNs are perhaps not valid for

UWSNs. So, an overall examination of the total network construction is necessary to

supply a suitable network service for observing the system such as routing and MAC

protocols, range of each node, network topography and all network necessities.

On the other hand and after the arrival of data from the wireless sensor

network; it will be processed. There is an important challenge to assess this data,

make it ready, and pass it to the concerned authorities. This is useful to monitor

environmental pollution in this area. For example, if the water quality is good,

weak, average or bad etc. Then, make the right decision to deal with this condition

of water. The second part is not essential, but very important in the completion of

an integrated pollution monitoring system. There is no need to bring other systems

that might complicate it. Water quality assessment is an evaluation of the water

body conditions using biological surveys, chemical-specific analyses of pollutants

in water bodies, and toxicity tests.

There are several ways to assess the quality of water using empirical equations

multiplying the indicators by certain weights depending on the degree of

importance. The latest method to assess the quality of water is the use of fuzzy

logic discovered by Lotfi Zadeh in 1965, which allows for using knowledge with


3

expertise in the processes to eliminate uncertainty and subjectivity contained in

traditional systems.

1.2 Research Motivation and Challenges

Two important concepts of research motivation and research challenges are

presented in this section. It introduces the traditional pollution monitoring research

motivation in its general form, then looks at the traditional way in Iraq. Challenges

are classified into two parts, challenges facing UWSNs and implementation of

networks.

1.2.1 Motivation

The old approach for monitoring is to allocate underwater sensors that record

data during the monitoring task, then mend the instruments. This approach has the

following disadvantages [4]:

1. No immediate monitoring: The recorded data can not read until the instruments

are mended. This may last for days, weeks, or months after starting the

monitoring task. In observation or environmental monitoring applications such

as seismic monitoring, instantaneous data retrieval is crucial.

2. No connected system reconfiguration: Collaboration between onshore control

systems and the monitoring devices are not possible. This does not obstruct any

adaptive regulation of the devices; nor is it possible to reconfigure the system

after specific events happening.

3. No failure discovery: If failures or misconfigurations happen, it might not be

possible to be discover them before the devices are recovered. This can easily

lead to the broad failure of a monitoring task.

4. Inadequate storage capacity: The quantity of data that can be recorded during the

monitoring task by every sensor is limited by the capacity of the aboard storage

devices (memories, hard disks).


4

These disadvantages of old underwater monitoring techniques bound the possible

applications for this atmosphere. Alternatively, the distributed sensor network model

may provide capabilities meaningfully exceed the existing underwater applications.

Therefore, there is a necessity to position underwater networks that enable

immediate monitoring of particular ocean areas, distant configuration and interaction

with onshore human operatives. All this can be gained by joining underwater

instruments by means of wireless links and forming underwater sensor networks.

In Iraq, and through a questionnaire conducted in a number of environment and

health centers involved in the control of pollution in the water, especially those

located in remote places and it was found that the monitoring process routine used is

a process tired, slow and contains all the disadvantages mentioned above. The

questionnaire to summarize system specifications approach is shown in Table (1.1).

1.2.2 Research challenges

The research challenges are classified into three categories; UWSN, network

implementation, and traditional water quality index challenges.

1.2.2.1 UWSN challenges

The major challenges encountered in the design of underwater acoustic

networks are listed as follows [5], [6] and [7]:

1. It is necessary to develop less expensive, robust nano-sensors, e.g., sensors

based on nanotechnology, which contains development of supplies and systems

at the nuclear, molecular, or macromolecular ranks in the dimension range of

about 1–500 nm.

2. It is essential to develop the mechanisms of scheduled by cleaning against

weathering and fouling, which may affect the lifespan of underwater devices.


5

Table (1.1): Questionnaire in five pollution monitoring centers

Questions Answers

The difficulty of monitoring process (Easy, difficult, very difficult)

Difficult

Number of times for sampling 6 times

The actual number of sampling 3 times

Is there a taking of samples in holidays No

Average number of employees in each center

2-5 employers

Time taken for sampling 2-4 hours

Constraints of this method documentation, transportation,

coordination with other agencies,

cost and other things

Authority interaction Slow and routine may take one

week

3. There is a need for healthy, constant sensors on a high range of heats since

sensor drift of underwater devices may be a concern. For this, protocols for in

situ calibration of sensors to increase correctness and exactness of sampled data

must be established. There is a necessity for new combined sensors for synoptic

sampling of physical, chemical, and biological parameters to develop the

accepting of processes in marine systems.

4. The available bandwidth is severely limited.

5. The underwater channel is impaired because of multi-path and fading.


6

6. Propagation postponement in subsurface is five orders of magnitude higher than

in Radio Frequency (RF) terrestrial channels, and variable.

7. High bit fault amounts and short-term losses of connectivity (shadow zones)

can be practiced.

8. Underwater sensors are categorized by high cost because of extra protecting

sheaths needed for sensors and also rather small numbers of suppliers (i.e., not

much reduced of scale) are available.

9. Battery power is restricted and usually batteries cannot be recharged, as solar

energy cannot be broken.

10. Underwater sensors are more disposed to failures because of entangling and

corrosion.

1.2.2.2 Network Implementation challenges

The difficulty and cost of providing professional laboratory for a real test lead

to the use of simulators. The major challenge is to find a suitable simulator to offer

all or most of the requirements of this underwater network. The challenges

encountered in the application of this network and the differences from those

worldly networks are:

1. audio communication is the usual conventional method for underwater

environments, while the broadcast speed of audio signal under water is very slow

(about 1500 m/s), considerably different from that of radio signal.

2. The audio signal decrease model is radically different from that of radio signal,

and thus audio channel models should be unified.

3. Underwater sensor networks are usually positioned in a three-dimensional place,

while these simulators usually only support two-dimensional placement. Thus,

the unique features of underwater sensor networks make the existing network

simulators unsuitable.


7

1.2.2.3 Traditional water quality index challenges

The use of these traditional indices raises many challenges:

1. Some parameters in the index equations can influence dramatically the final

score without valid justification, while their formulations are rather elementary,

and the number of variables involved is too limited. [8].

2. This leads to an unclear distinction between each mode of the index and

causes inaccuracies and ambiguity when making decisions about boundary

values [9].

3. The most critical deficiency of these indexes is the lack of dealing with

uncertainty and subjectivity [10].

1.3 Literature Review

The related works of wireless sensor networks simulation are discussed in

this section.

In [11], the author proved that low-cost sensors are usable in underwater

sensor networks and the Microsoft Windows CE may be used in the Intel Xscale

PXA270 board for signal processing. A high-level programming language like

C++, C# or VB.Net used for programming the applications that read real-time

sensor data and convey sensor data between a sensor node and a super-node. He

presented that the sensor data can be read by the Intel Xscale PXA270 board via

its analog to digital (A/D) converter. The sensor data can be conveyed from a

sensor node to a super-node in real time throughout an 802.11 wireless sensor

network

In [12], an original alternative opportunity of time-critical underwater

communication using short-range (50 – 500 m) low-cost sensors was presented.

Their chief goal is that the modem must be cheap to make it possible to buy and

position many underwater sensor nodes. Multi-hop routing over many separate

nodes can reach long-range communication. Actually, concentrating on short-


8

range communication means can increase the obtainable audio bandwidth and

also escape many of the challenges of long-range underwater communication

and henceforth, really make simpler the modem design. In addition, the

researcher has planned a novel management protocol that confirms total volume

coverage, advance the QoS; in addition to enhance the lifetime of the entire

network.

In [13], the authors introduced the adjacent of network nodes form

cluster.in which, each node sends the data to the main node of the cluster. The

main node compresses the data and sends it to the Sink node. Actually, sink

node is the gateway nodes, responsible of the network beginning, repairs, data

gathering and send data to the control center. The monitor center is responsible

of data processing and network management. There is some specific software on

the control center that does the job of data processing and makes choice. As

woodland is a place that human can easily reach, so artificial split farmland can

be split into multiple regions, each region is a cluster of network topology. And

inside each cluster a head node responsible of the communication with gateway

is assigned. And meanwhile the agricultural environment may not have standard

cable network, so they are considered as two communication structures. (a) The

gateway communicates with the Server control center through the cable

network. (b) In the mobile networks such as Global System for Mobile

Communications (GSM) or Code division multiple access (CDMA) coverage

area can be used as a broadcast medium. The sink node sends the data to main

stations, and the main station data is then transmitted to the monitoring center.

In [14], the authors explained and illustrated a new design of water

environment monitoring system, based on a WSN. The system generally

includes three divisions: hardware and software of data monitoring nodes,

hardware and software of the data main station, in addition to software for the

distant monitoring center. The system efficiently achieves an on line auto


9

monitoring of the water temperature and pH environment of a non-natural lake.

Sensors were appropriate to alter water quality that can be installed at the node

to meet the monitoring requirements in various water environments and to attain

various parameters.

In [15], the authors developed and compared different sensor network

construction designs that can be used for monitoring underwater pipeline

foundations. These constructions were underwater wired sensor networks,

underwater audio wireless sensor networks, RF (Radio Frequency) wireless

sensor networks, combined wired/audio wireless sensor networks, and integrated

wired/RF wireless sensor networks. The researchers also discussed the

dependability challenges and enhancement methods for these network

constructions. The dependability evaluation, features, compensations, and

shortcomings among these constructions were argued and matched. Three

reliability factors were used for the discussion and comparison: the network

connectivity, the endurance of power supply for the network, and the physical

network security. In addition, they developed and evaluated a graded sensor

network framework for underwater pipeline monitoring.

In [16], Based on wireless sensor networks (WSN), reported a new

observation system using underwater multisensory information. After processing

the multisensory data from each sensor, the system transmits it to a hub node

through WSN, and then transmits it to a land data center through a general

packet radio service (GPRS) wireless network. In order to check the basic

performance of this system, they completed a node positioning experiment

based on a GPS module, and a communication experiment based on ZigBee.

This article reported the design of the hardware and the experimental results

[16].


10

In [17], the author introduced a thesis focused on node lifetime extension

based on energy management. While some constraints and results might hold

true from a more general perspective. The main application target involves

environmental measurement systems based on Wireless Sensor Networks.

Lifetime extension possibilities, which were the result of application

characteristics, by (i) reducing energy consumption and (ii) utilizing energy

harvesting were to be presented. For energy consumption, he showed how

precise task scheduling due to node synchronization, combined with methods

such as duty cycling and power domains, can optimize the overall energy use.

With reference to the energy supply, the focus lies on solar-based solutions with

special attention placed on their feasibility at locations with limited solar

radiation. Further dimensioning of these systems was addressed [17].

In [18], the authors developed an underwater sensor network for

monitoring water quality and pollution to keep heat and pH as the parameters

which were being monitored by the jots and communicated through data main

station and GPRS modems to the data monitoring center. This can serve as a

prolonged solution for ecological control and monitoring of physical

environments therefore making pollution monitoring less complex and age

group of reports on fixed basis upholding a close check on the amount of

impurities through monitoring acidity in the samples of water under monitoring

[18].

In [19], the author presented a thesis about general framework in water

quality monitoring system. The real-time values of analytical instruments are

necessary to the wireless data acquisition terminal, the data processed and

packaged, and sent to the data center through a wireless network. After

decrypting, the system does the data analysis, storage, display and alarm

automatically by management information system (MIS) and geographic

information system (GIS). It then publishes the data to the upper network


11

control and management system Via TCP / IP protocol and exchange data with

other control centers. At last, the center sent the command to the sub-stations

and summaries the feedback through GSM/GPRS communication. In short

range, Zigbee and Wi-Fi frequently are used in water monitoring area. For

sensor cost wireless networking protocol are targeted towards automation. On

the other hand, GSM/GPRS are applied for long range communication [19].

In [20], the authors identified a need for an autonomous and

collaborative mechanism, based on targeted wireless sensor technologies.

Furthermore, they concluded that for effective and integrated water quality

monitoring and management at a catchment scale, a system of individually

networked activities is needed. This is likely to take the form of a combination

of networks that have been installed for local monitoring purposes (e.g. at farm

scale, primarily to provide local information for that farm), and some specific

networks aimed at filling in information gaps in the catchment. This system of

networks should be able to share information about critical parameters for

events, such as rain, or floods, that could trigger consequences, such as

contaminants runoff. It requires a higher level application to make use of this

information, and that the individual networks are aware of similar networks

nearby capable of passing information to them. This inevitably led them to some

form of standardization of communication protocol and data representation

between enabled environmental networks [20].

By looking carefully at the pollution monitoring systems mentioned

above, the following disadvantages are noted:-

1. It can be seen that the above researches did not contain physical comparison

between the types of communications in terms of facing the UWSN challenges

then select the appropriate one. For example the comparison between acoustic,

electromagnetic and optical communication.


12

2. The lack of a comparative study between MAC and routing protocols to find the

most appropriate one. This comparison is done through specific metrics such as

(energy consumption, received throughput, packet delivery ratio, total drop

packet and etc.).

3. The number of nodes used is not taken into consideration the possible number

that is optimum to get an efficient system by using specific metrics.

4. Lack of a comprehensive study on the importance of the pollutants to be

monitored, is it sensitive information or not? Can it be delaying or not? And

other then decide on any routing protocols will use depending on type of the

system.

5. No focus on important aspect especially, Integrated Systems in terms of the

design of monitoring systems, data processing, water quality assessment, make

decision when there is an increase or decrease in limits of indicators and

reporting to the competent authorities.

This is evident through the research, conducted in a number of scientific

papers related to the terms "Monitoring", "underwater wireless sensor network"

and "ns2". There are 1250 papers on this matter. While for the terms

"Monitoring", "underwater wireless sensor network", "ns2" and "fuzzy water

quality" only 12 papers were found that simply deal with these concepts as an

integrated system.

1.4 Aim of work

The aim of presented work is to overcome the gaps in literature mentioned

above and obtain preferred and integrated monitoring system.

1. These works is design and implement the UWSN to monitor some of the

pollutants in the Iraqi rivers.

2. Through this design, several comparisons between communication channels,

MAC protocols and routing protocols is presented to obtain the results for most

appropriate one for this work from the point of view of several metrics.


13

3. To develop a novel water quality index in Iraq based on fuzzy inference system,

that is, a comprehensive artificial intelligence (AI) approach to the development

of traditional environmental indices for routine assessment of water quality,

particularly for human drinking purposes.

1.5 Thesis Layout

This thesis is organized to 5 chapters. Chapter 2 presents the overview of

UWSN as architecture of sensor, communication medium, MAC protocols and

routing protocols including a quick comparison between the terrestrial WSN and

UWSN. It additionally presents a compaction between the present simulators and

candidate the appropriate one to face the implementation challenges. It also

presents a background about water quality indices and presents the material tools

for developing the traditional way used in Iraq used to assess water quality.

Chapter 3 presents Performance Assessment of MAC Layer Protocols in UWSN

by using the specific metrics (energy consumption, received throughput, total drop

packets and average end to end delay) and elects the appropriate one based on

present system requirements and the UWSN routing protocol. It also conducts a

comparative study of them in terms of energy consumption, average end-to-end

delay and Packet delivery ratio to choose the most appropriate. Additionally,

Chapter 3 provides the final evaluation of the system in terms of the metrics that

are adopted.

Chapter 4 presents the details of the traditional way to assess water quality and

compare it with a novel way in Iraq that use fuzzy inference system. The use of

FIS step by step, the development of index and result of compression will also be

introduced in this chapter, together with the validation of the new fuzzy index.

Chapter 5 summarizes the thesis and conclusion with a suggestion of future work

trends.


14

Chapter Two

Theory of Pollution

Monitoring

Systems

Chapter Two ….…………………….…… Theory of Pollution Monitoring Systems

14

Chapter Two

Theory of Pollution Monitoring Systems

2.1 Introduction

In this chapter, the theoretical background of the pollution monitoring system

is discussed. This background contains two categories. The first one is an

overview of UWSNs which include UWSN definition, differences between

terrestrial (normal networks) and underwater Sensor Networks, UWSN

applications, important comparison between all UWSN communication channels.

In addition, it describes in details all aspect of UWSN as components and

communication architecture. At last, the Protocol stacks for underwater acoustic

channels are presented with a highlight of MAC and Routing protocols.

The second one is water quality indices and the implementation material tools

and methods to develop the traditional way used in Iraq to assess water quality by

using fuzzy inference system.

2.2 Wireless Underwater Sensor Networks

Underwater sensor networks are planned to enable applications for

oceanographic data gathering, pollution checking, offshore examination, disaster

avoidance, aided navigation and strategic investigation applications [21].

2.2.1 Terrestrial Sensor Networks versus Underwater Sensor Networks

Several features of networking in water have correspondences with the

terrestrial sensor networks. There are many differences that need communication

protocols to be custom-made for underwater sensor networks. The

communication procedures and the variances are clarified in details in section 2.


15

2. Important differences between terrestrial and underwater sensor networks are

show in Table (2.1).

Table (2.1): Differences between terrestrial and underwater sensor networks [22].

2.2.2 UWSN Applications

From UWSNs features a broad range of underwater applications are [1]:

1. Environmental monitoring: Underwater sensor networks can accomplish

pollution monitoring (chemical, biological, and nuclear). For example, it is

possible to demonstrate the chemical slurry of antibiotics, estrogen-type

hormones, and insecticides to observer streams, rivers, lakes, and ocean bays

Cost Deployment Power Memory Spatial

Correlation

Terrestrial

Sensor

Networks

Inexpensive

devices

densely

deployed

Lower than

acoustic

underwater

communications

very

limited

storage

capacity

correlated

Underwater

Sensor

Networks

expensive

devices

generally

more

sparse

more power is

consumed in

comparison

with terrestrial

do some

data

caching as

the

underwater

channel

may be

intermittent

uncorrelated


16

(water quality in-situ examination) [23]. Monitoring ocean streams and

winds, enhanced weather forecast, discovering climate change, understanding

and expecting the influence of human actions on naval ecosystems,

biological monitoring such as tracing fishes or microorganisms, are other

potential uses.

2. Undersea explorations: Underwater sensor networks can detect underwater

oilfields or reservoirs, determine routes for resting underwater cables, and

help in the investigation for valued minerals.

3. Assisted navigation: Sensors can be used to recognize threats on the seabed,

to locate dangerous rocks or shoals in shallow waters, mooring locations, and

flooded collisions, and to perform bathymetry summarizing.

4. Distrib uted tactical surveillance: AUVs and fixed underwater sensors can

cooperatively monitor zones for investigation, inspection, targeting, and

intrusion discovery systems. A good example, a 3-D underwater sensor

network can recognize a strategic investigation system that is able to sense

and categorize submarines, small delivery vehicles (SDVs), and divers built

on the sensed data from mechanical, radiation, magnetic, and audio micro

sensors. With respect to old-style radar/sonar systems, underwater sensor

networks can reach an advanced correctness, advanced coverage, and

forcefulness as well as allow discovery and cataloguing of low-signature

targets by joining measures from different kinds of sensors.

5. Mine reconnaissance: The concurrent action of various AUVs with audio

and optical sensors can be used to make quick environmental calculation and

discover mine-like objects.


17

2.3 Comparison between RF, Optical and Acoustic Underwater Communication

Effective underwater communication among units or nodes in a UWSN is

one of the utmost essential and serious issues in the entire network system

design [24].

Current underwater communication systems include the transmission of

information in the form of sound, electromagnetic (EM), or optical waves. All of

these techniques have benefits and boundaries. Audio communication is the

most multipurpose and broadly used technique in underwater settings because of

the low reduction of sound under water. This is particularly true in thermally

constant, deep-water environment. Alternatively, the use of audio waves in thin

water can be badly affected by heat rises, surface ambient noise, and multipath

broadcast because of reflection and refraction. The slower speed of audio

propagation in water, about 1500 m/s (meters per second), compared with that of

electromagnetic and optical waves, and forms another restrictive factor for

effective communication and networking. Nonetheless, the present promising

technology for underwater communication is upon audibility.

On the front of using electromagnetic (EM) waves in radio frequencies,

conventional radio does not work well in an underwater environment due to the

conducting nature of the medium, especially in the case of seawater. However, if

EM could be working underwater, even in a short distance, its much faster

propagating speed is definitely a great advantage for faster and efficient

communication among nodes.

Free-space optical (FSV) waves are used as wireless communication

carriers are commonly restricted to very small distances since the severe water

absorption is at the optical frequency bands that produce strong backscatter from

hanging particles. Even the purest water has 1000 times the reduction of clear

air, and muddled water has more than 100 times the reduction of the heaviest


18

fog [24]. Underwater FSV, specifically in the blue-green wavelengths,

compromises a real choice for high-bandwidth communication (10-150 Mbps)

over reasonable ranges (10-100 meters). This communication range is much

desired in port examination, oil-rig repairs, and connecting submarines to land,

these are just names of few of the demands on this front [24].

Table (2.2) summarizes and compares the characteristics of radio, optical,

and acoustic communication underwater. All three physical wave fields have

their own advantages and limitations for acting as an underwater wireless

communications carrier; radio waves can provide high data rates, but are subject

to strong attenuation by the conductivity of sea water, optical waves provide

even higher data rates, but are subject to attenuation by the turbidity of sea

water, acoustic waves provide long transmission distances but support relatively

low data rates and are subject to multipath. As our sensor network applications

require low data rates and transmission distances greater than 100 meters,

acoustics remains the most robust and feasible carrier to date for wireless

communication in these underwater sensor networks. As acoustics have been

widely used in underwater communications and we have selected acoustics for

our monitoring system.

2.4 Underwater Sensor Network Components

The design difficulties and the sole features of underwater sensor networks

need different components for the understanding of these networks. In this

section, we designate these apparatuses of the underwater sensors that are used

to gather information about the underwater setting.


19

Table (2.2): Comparison or RF, optical and acoustic communication underwater

[25].

The typical internal architecture of an underwater sensor is shown in Figure

(2.1). It consists of a main controller/CPU which is interfaced with an

oceanographic instrument or sensor through sensor interface circuitry. The

controller receives data from the sensor and can store the data in the on-board

memory, process them, and send them to other network devices by controlling

the acoustic modem. The electronics are usually mounted on a frame which is

protected by polymerizing vinyl chloride (PVC) housing [26].

RF Optical Acoustic

Wave speed

(m/s)

~3E8 ~3E8 ~1.5E3

Data rate < 10 Mbps <1 Gbps < 100 Kbps

Effective range ~ 1-100 m ~ 1-100 m ~ 1km

Power Loss ~ 28

dB/1km/100MHz

α turbidity > 0.1 dB/m/Hz

Frequency

Band

~ 1MHz ~104 – 1015Hz ~1kHz

Major hurdles power limited environment

limited

bandwidth

limited

interference-

limited


20

Figure (2.1): Internal organization of an underwater sensor node [2].

Occasionally all sensor apparatuses are secured by bottom-mounted

instrument frames that are designed to license azimuthally omnidirectional audio

communications, and keep sensors and modems from the possible influence of

seeking gear, particularly in zones subject to fishing actions.

In addition to the main controller and the transceiver circuitry, underwater

sensor devices are equipped with a vast variety of sensors. These devices

contain sensors to measure the quality of water and to study its features such as

heat, density, salinity (interferometry and refract metric sensors), acidity,

chemicals, conductivity, pH (magneto flexible sensors), oxygen (Clark-type

electrode), hydrogen, dissolved methane gas (METS), and turbidity. One type of

the used sensors is that used to discover ricin– the highly toxic protein found in

castor beans and thought to be a possible terrorism agent. Deoxyribo Nucleic

Acid (DNA) microarrays can be used to monitor both abundance and activity-

level variants among natural microbial inhabitants. Other existing underwater

sensors include hydrothermal sulfide, silicate, voltammetry sensors for

ACOUSTIC MODEM

SENSOR INTERFACE CIRCUITRY

MEMORY

SENSOR

CPU-ONBOARD CONTROLER

POWER SUPPLY


21

spectrophotometry, gold-amalgam electrode sensors for sediment measurements

of metal ions (ion-selective analysis), aerometric micro sensors for H2S

capacities for studies of an oxygenic photosynthesis, sulfide oxidation, and

sulfate reduction of sediments [2].

2.5 Communication Architecture

In this section, we describe the communication architectures of underwater

acoustic sensor networks. In particular, we introduce reference architectures for

two-dimensional and three-dimensional underwater networks.

Underwater monitoring missions can be extremely expensive due to the high

cost of underwater devices. Henceforth, it is significant that the positioned

network be highly dependable, to avoid failure of monitoring tasks due to failure

of single or multiple devices. For example, it is vital to evade designing the

network topology with sole points of failure, which could compromise the total

operational of the network. The network capability is also affected by the

network topology. Since the capacity of the underwater channel is severely

limited it is very important to organize the network topology in such a way that

no communication bottleneck is introduced [27].

2.5.1 Two-dimensional Underwater Sensor Networks

Position construction for two-dimensional underwater networks is shown

in Figure (2.2). A group of sensor nodes is attached to the bottommost of the

ocean with deep ocean anchors. Underwater sensor nodes are interrelated to one

or more underwater gateways (uw-gateways) via wireless audio links. UW-

gateways, as presented in Figure (2.2). These are network devices responsible of

spreading data from the ocean bottom network to a surface station. To attain this

goal, uw-gateways are prepared with two audio transceivers, namely a vertical

and a horizontal transceiver. The horizontal transceiver is used by the uw-


22

gateway to communicate with the sensor nodes to: i) direct instructions and

shape data to the sensors (uw-gateway to sensors); and ii) collect monitored data

(sensors to uw-gateway). The vertical link is used by the uw-gateways to relay

data to a surface station. In deep water applications, vertical transceivers must be

long range transceivers as the ocean can be as deep as 10km. The surface station

is prepared with audio transceiver [2], [9] and [27].

Figure (2.2): Architecture for 2D underwater sensor networks [2].


23

2.5.2 Three-dimensional Underwater Sensor Networks

Three dimensional underwater networks are used to distinguish and notice

sensations that cannot be sufficiently noticed via ocean bottom sensor nodes,

i.e., to achieve supportive sampling of the 3D ocean setting. In three-

dimensional underwater networks, sensor nodes drift at various depths to notice

a particular phenomenon. One probable solution would be to attach each uw-

sensor node to a surface marker, via wires whose length can be controlled so as

to adjust the depth of each sensor node.

Figure (2.3): Architecture for 3D underwater sensor networks [2].

Though, this answer permits easy and hurried placement of the sensor

network, multiple detached markers may block ships crossing on the surface, or

they can be easily discovered and disabled by foes in military surroundings.


24

Additionally, floating markers are defenseless to weather and tampering or

stealing. For these details, a different way can be used to anchor sensor devices

to the bottom of the ocean. In this construction, depicted in Figure (2.3), each

sensor is anchored to the ocean bottom and prepared with a floating marker that

can be inflated by a pump. The marker pushes the sensor in the direction of the

ocean surface. The depth of the sensor can then be controlled by correcting the

length of the wire that attaches the sensor to the anchor, via an electronically

controlled engine that exists on the sensor. A challenge is faced in such

construction that is the effect of ocean flows on the described mechanism to

regulate the depth of the sensors.

2.6 Protocol stack for underwater acoustic channels

In general, the protocol mass used by the sink and all sensor nodes is

given in Figure (2.4) [2]. This protocol mass combines power and transmitting,

incorporates data with networking protocols, transfers power effectively over the

wireless medium, and supports collaborative efforts of sensor nodes. The

protocol mass involves the physical layer, data link layer, network layer,

transportation layer, application layer, in addition to synchronization plane,

localization plane, topology management plane, power management plane,

mobility organization plane, and task organization plane. The physical layer

reports the requirements of simple but vigorous modulation, transmission, and

receiving techniques. As the setting is noisy and sensor nodes can be moveable,

the link layer is responsible of ensuring dependable communication over fault

control techniques and manages channel access over the MAC to reduce impact

with neighbors’ broadcasts. Depending on the sensing jobs, various kinds of

application software can be made and applied on the application layer. The

network layers maintain routing the data provided by the transportation layer.

The transportation layer aids to keep the stream of data if the sensor network

application needs it. Furthermore, the power, flexibility , and mission


25

organization planes monitor the power, movement, and mission spreading

among the sensor nodes. These planes aid the sensor nodes manage the sensing

mission and decrease the total power intake. The power organization plane

accomplishes how a sensor node uses its power. For example, the sensor node

may shoot off its receiver after receiving a message from one of its neighbors.

This is to avoid getting duplicated messages. Also, when the power level of the

sensor node is low, the sensor node broadcasts to its neighbors that it is small in

power and cannot contribute in routing messages. The remaining power is kept

for sensing. The flexibility organization plane discovers and records the

movement of sensor nodes, so a route back to the user is always preserved, and

the sensor nodes can preserve track of their neighbors. By knowing these

neighbor sensor nodes, the sensor nodes can set of scales their power and task

treatment. The mission organization plane balances and schedules the sensing

jobs specified to a particular area. Not all sensor nodes in that area are necessary

to achieve the sensing job at the same time. Accordingly, certain sensor nodes

achieve the job

more than others, due

to their power level.


26

Figure (2.4): The sensor network protocol stack [2].

These organization planes are required so that sensor nodes can work together in

a power-effective way, route data in a moveable sensor network, and share

resources between sensor nodes. Without them, every sensor node will just work

alone. From the position of the entire sensor network, it is more effective if

sensor nodes can cooperate with each other, so the lifespan of the sensor

networks can be extended.

The next two sections present a detailed explanation of MAC and routing layer

protocols in UWSN. It is considered that the most important challenge facing

the network is this layer.

2.6.1 MAC Layer Protocols

In UAWSN, MAC constitutes one of the major challenges in sensor

networks [28]. Such as available bandwidth is severely limited, Propagation

delay in underwater is five orders of magnitude higher than in Radio Frequency

(RF) terrestrial channels and Battery power is limited and usually batteries

cannot be recharged as solar energy cannot be exploited [29].

The main task of MAC protocols is to provide efficient and reliable access

to the shared physical medium in terms of throughput, delay, error rates and

energy consumption [30]. However, several drawbacks are faced with the

suitability of the terrestrial MAC solutions for the aquatic environment, because

of the different nature of the underwater environment. The Frequency Division

Multiple Access (FDMA) is narrow bandwidth available in underwater acoustic

channels, and the vulnerability of limited band systems to fading and multipath

effects. Also, Time Division Multiple Access (TDMA) shows restricted


27

bandwidth efficiency because of the long time guards required in the underwater

acoustic channel [31].

There are two categories suitable for UAWSN as Carrier Sense Multiple

Access (CSMA) and Code Division Multiple Access (CDMA). In general,

CSMA-based protocols are vulnerable to both hidden and exposed terminal

problems. In order to decrease the effects of hidden terminals MAC proposals

should include techniques similar to those used in terrestrial networks like

MACA [32], which uses RTS/CTS/DATA packets to reduce the hidden terminal

problem. And MACAW [33], which adds to the previous one an ACK packet at

the link-layer that can be profitable in an unreliable underwater channel. FAMA

[34] extends the duration of RTS and CTS packets so as to avoid data packet

collisions, thus, contention is managed at both sender and receiver sides before

data packets are sent. The efficiency of these protocols is heavily impacted by

propagation delays due to their multiple handshakes. CDMA-based protocols

are not useful for acoustic networks because these protocols have some

problems such as synchronization and near far problem [28].

Underwater MAC layer protocols should also assume node mobility, low

bandwidth, energy efficiency and long propagation delay. Due to the long

propagation delay, node mobility and other underwater environment constraints,

distributed topologies are used more than centralized topologies. Thus,

contention based protocols like Broadcast MAC, Aloha; R-MAC, FAMA, and

UWAN_MAC are useful for such topologies.

2.6.1.1 Floor Acquisition Multiple A ccess (FAMA)

FAMA is proposed by Chane and Garcia [34]. The objective of a FAMA

protocol is to allow a station to acquire control of the channel (the floor)

dynamically, and in such a way that the data packets never collide with any

other packet. This can be viewed as a form of dynamic reservations; however, in


28

contrast to prior approaches to dynamic reservations, which are also referred to

collision avoidance schemes the FAMA protocols presented requires no separate

control sub-channels or preambles to reserve the channel. Instead, a FAMA

protocol involves a station who wishes to send one or more packets to acquire

the floor before transmitting the packet train. The floor is acquired using control

packets that are multiplexed together with the data packets in the same channel

in such a way that, although control packets may collide with others, data

packets are always free of collisions.

A floor acquisition strategy based on an RTS-CTS exchange is particularly

attractive in the control of packet- radio networks due to its ability to provide a

building block to solve the hidden-terminal problem that arises in CSMA.

Within the context of using an RTS-CTS exchange for floor acquisition, there

are many ways in which such control packets can be transmitted. Two variants

are interest in this work; namely:

• RTS-CTS exchange with no carrier sensing.

• RTS-CTS exchange with non-persistent carrier sensing.

The first variant corresponds to using the ALOHA protocol for the

transmission of RTS packets. While the second is consists of using the non-

persistent CSMA protocol to transmit RTS packets. We choose to consider non-

persistent carrier sensing over persistent carrier sensing, because the throughput

of non-persistent CSMA is much higher under high load and only slightly lower

under low load than the throughput of p-persistent CSMA. Despite that the

original motivation for MACA was to solve the hidden-terminal problem of

CSMA, the basic RTS-CTS dialogue of MACA and even a four way handshake

(RTS, CTS, data, acknowledgment) does not solve all hidden- terminal

problems. For example, as Figure (2.5) shows, three given stations S, R and H.

If H is “hidden” from S (i.e., S and H cannot hear each other's transmissions) it


29

could happen that S sends an RTS to R in the clear and R sends a CTS to S. The

problem occurs when H transmits an RTS to R, or another station that can hear

R and H, at the same time that R transmits its CTS to S. If this is the case, then S

will send data packets to R, and H may transmit an RTS that R can hear and

collide with S’s data packets. Obviously, an ad-hoc solution would make H wait

a very long time before trying to retransmit, but that would degrade the network

throughput. The four-way handshake advocated in the IEEE 802.11 only helps

detecting hidden-terminal interference after it occurs, but, does not prevent it.

The RTS-CTS dialogue can be used as the building block to reduce the

hidden-terminal problem. However, this work focuses only on using such a

dialogue to establish a floor acquisition discipline, and focuses on single-hop

networks in which hidden terminals does not exist. The design of FAMA

protocols for multi-hop packet-radio networks is addressed elsewhere. The basis

for such protocols is the use of additional feedback from the receiver, in the

form of CTSs and partial acknowledgments to packet trains.

The drawbacks of FAMA are the following [36]:-

- Difficulty in configuring FAMA protocol when a new node come in or

moves out of network system.

- It is hard to implement FAMA protocol in a distributed mode.

2.6.1.2 Routing enhanced MAC (R-MAC )

The major design objectives of R-MAC are energy efficiency and fairness.

R-MAC schedules the transmissions of control packets and data packets to avoid

data packet collision completely. The scheduling algorithms does not only save

energy but also solve the exposed terminal problem inherited in RTS/CTS-based

protocols.


30

Moreover, the scheduling algorithms allow nodes in the network to select

their own schedules, thus loosening the synchronization requirement of the

protocol. Additionally-MAC supports fairness. R-MAC is divided into three

phases. In the initial phase (Latency Detection), nodes measure the distances to

neighbors by sending some small control packets. Based on the measurements,

every node will randomly choose a period for its data transmission and inform

others in the second Phase (Period Announcement).

In Phase 3(Periodic Operation), nodes cooperate with each other to

schedule data transmissions to avert collisions [35]. R-MAC is a fair MAC layer

protocol. Such that an intended receiver can provide equal opportunities to make

reservation for all its neighbors using REV and ACK-REV packets. This

protocol is good when there is no new node joins in the network and that all the

nodes are static.

The main drawback of R-MAC is that there is no technique can be proposed

for the node which tends to change its transmission schedule, or when a node

fails or a new node joins are the network [37].

Figure (2.5): Hidden terminal example: The transmission from R and H collide

at t2, leading to collision of S's data with H's RTS.


31

2.6.1.3 Underwater Acoustic MAC (UW-MAC)

UWAN-MAC [38], also deploys CSMA-based MAC and has been

primarily developed for high density UWSNs. Rather than bandwidth

optimization, UWAN-MAC focuses on energy efficiency by introducing sleep

schedules similar to its terrestrial counterparts. Each node has a sleep schedule

such that each node wakes up periodically in the network to transmit its data. At

the beginning of each cycle, a node broadcasts a SYNC packet indicating its

period of the sleep schedule. And as a result, the neighbor nodes that receive this

packet wake up at the next scheduled time to listen to the node. Consequently,

every node wakes up for each of its neighbors to receive data in addition to its

scheduled wakeup time to transmit data. Note that since relative time

information is exchanged by the SYNC packets, UWAN-MAC propagation

delay is not required to be known by other nodes.

The operation of the UWAN-MAC synchronization mechanism is shown

in Figure (2.6). When node a broadcasts a SYNC packet; it indicates its sleep

period as TA. Accordingly, when node A’s neighbors receive this SYNC packet,

they schedule to wake up TA seconds after reception of the SYNC packet.

Similarly, node A also receives SYNC packets from its neighbors and schedules

wakeup times for them. The data transmission packet structure of each node is

shown in Figure (2.7), which consists of Missing, SYNC, Data Tx, and Listen

periods. The SYNC period is used to broadcast SYNC packets as explained

before, while Data Tx is used to transmit the DATA packets. Since each of the


32

neighbors of node A is listening to the transmission period of the node, it can

transmit its DATA without any collisions. The missing and listening periods are

used to handle node failures/removals and node joins. At each sleep period, a

node collects the list of its own neighbors that it has received SYNC messages

from. In case that there is a change in this list (a SYNC

message from a certain node is not received), the node creates a missing node

list and broadcasts this information during the Missing period shown in Figure

(2.7). This list serves as notification to the nodes in the missing list that a

communication error may have occurred earlier. If a node does not hear from its

neighbors in the missing list for a couple of consecutive cycles, it deletes this

node from its neighbor list. On the other hand, the node that is in the missing list

replies back to the sender of the SYNC message as if it is a newcomer node. The

procedure for newcomer nodes is explained next.

The listening period in the transmission period shown in Figure (2.7) is

used to comprise newcomers to the network. This situation is illustrated in

Figure (2.6) where node D joins the network while node C is transmitting a

SYNC packet. When node D joins the network, it listens to the channel for the

SYNC packets from its neighbors. When it receives a SYNC packet from node

C, it makes a reply to this packet with a HELLO packet to indicate its existence.

The Listen period at the end of each transmission period ensures that node C

receives this HELLO packet. Then, node C includes the newcomer node D in its

list of neighbors. In the HELLO packet, node F also indicates the time left for its

next wakeup time, i.e. Node C can then wake up for the scheduled wakeup time

of node D and receive its SYNC packet as shown in Figure (2.6) Node D

indicates its schedule to other nodes in the same manner [38].

The operation of UWAN-MAC so far assumes that the propagation delay

between two nodes does not change. This enables the relative wakeup

announcements by the SYNC packets to synchronize nodes. However, the


33

underwater acoustic channel suffers from high variable propagation delays and

channel fluctuations and that is consequent to many reasons such as the drifts of

nodes, scattering objects in the water and multi-path effects. And thus, the

resultant propagation delay fluctuates randomly. So, to account for this

fluctuation, UWAN-MAC are introduces guard times before and after each

listen duration of its neighbors. This ensures that a packet is correctly and fully

received even if it arrives earlier or later than expected. An implicit problem

with this extension is that a node’s transmission schedule and its own listen

period may overlap because of the guard times. In such case, the node changes

its transmission schedule and broadcasts this via a SYNC packet [38].

UWAN-MAC achieves significant energy consumption through the

sleep schedules. Since it is developed for the applications of delay-tolerant, the

sleep schedules may induce very high medium access delays for

communication. Furthermore, the throughput of the protocol is decreased due to

the overhead in maintaining schedules and the sleep schedule operation [38].

Figure (2.6): The operation of the UWAN-MAC mechanism [38].


34

Figure (2.7): The transmission packet structure and the listen duration for the

UWAN-MAC mechanism [38].

2.6.1.4 Aloha

This protocol is based on the Aloha idea while tailored to underwater

network environments: when a node has packets to send, it will send them

directly without sensing the channel. The sender then will start a timer and waits

for the response from the receiver. If the receiver receives a packet correctly, it

will send an acknowledgment (ACK) back to the sender. If the sender receives

an ACK before it time out, the sender knows that this packet has been

successfully transmitted and starts to send the next packet. Otherwise, the sender

will back off for some time and resend the same packet again.

2.6.1.5 Broadcast MAC

When a node has packets to send, it first senses the channel. If there is no

transmission at a time, it broadcasts the packets. Otherwise, it backs off. Packets

are dropped if the number of back-off times exceeds the limit. When the receiver

receives a packet, there is no need to send an ACK back to the sender. This

protocol is simple yet efficient in low traffic networks. In addition, this protocol

can take full advantages of the broadcast nature of the underwater acoustic


35

channel and are suitable for geo-routing protocols such as vector based forward

(VBF) [39].

2.6.2 Routing protocols

The network layer function is finding a way from source to the destination

taking into account many characteristics of the channel such as long propagation

delay and energy of the nodes. There was an extensive study to find the path

from the source to the destination in various gateways of the UWSNs. These

protocols can be classified into three different groups: proactive, reactive and

geographical routing [28]. From [40], [41] and [42] an appropriate proactive and

reactive protocol with UWSNs is too weak, for memory, energy reasons and

incompatibility of proactive protocol with UWSN. Reactive protocols are

unsuitable for underwater networks because of high latency, asymmetrical links

and topology and so the higher delay to create the path, being further amplified

in this environment because the slower propagation in acoustic signals. Thus,

the geographical routing protocols are energy efficient and scalable. For these

reasons the most suitable approach using in UWSNs is geographical routing

protocols. A brief on some routing protocols designed to UWSNs topologies

will presented later. Most of them take into consideration the limitation in

energy [43].

2.6.2.1 Vector based forward (VBF)

VBF is robust, scalable and energy efficient [44]. This is mainly in "routing

pipe" approach. There is no need for Information Service on the nodes, except

only a small part of them. Also, the packets pass through repeated and

dovetailed paths from source to sink; therefore VBF is robust against losing

packets and frailer can occur on nodes.

The routing in VBF routing protocol


36

In VBF, each packet holds the location of the sender (SP), the destination

(TP) and forwarder (FP). Each packet contains a RANGE field. The packet that

reaches the area defined by its TP, that packet is controlled by the RANGE field.

The routing pipe is the vector from the sender (SP) to the destination (TP) and

the radius of the pipe is illustrated in the RADIUS field. Routing in VBF is

embarked by query packets. VBF routes various queries in deferent ways [44]:

i. Sink Initiated Query :-

Two types of such queries: location-dependent query, in this type the sink is

interested in some limited area and knows the position of that area. The other

type of this query is location-based query, in which the sink needs to know some

types of data or information regardless of its position.

ii. Source Initiated Query:-

When a source is starting to transmit, firstly sets a coordinate system for

itself and floods a packet called "DATA READY" to the network. Thus, each

node and sink can calculate its position in the source-based coordinate system.

The sink sends source location to its coordinate system, and sends a location-

based packet to source to give it permission for computing its location in the

sink-based coordinate system for following communications.

VBF Algorithm

A node receiving packet calculates its position and checks if it is in the

routing pipe. If so; the node holds the packet for a time called Tadaptation. This

is determined as follows [44]:

Tadaptation = √α × tdelay + (r-d/υ0) (2.1)


37

Where α is the desirableness factor, tdelay is a maximum delay, υ0 is the

propagation speed of acoustic signals in water (1500m/s), and d is the distance

between node and the forwarder. The theoretical analysis can be found in [45].

2.6.2.2 Hop-by-Hop Vector-Based Forwarding (HH-VBF)

Drawbacks of VBF

There are two main problems in VBF routing protocol [46]:

• The data has a very small deliver ratio for dispersed networks, if nodes

fall under these pipes, the data packets can't be forwarded to the sink until

paths may exist out of pipe. These paths do not exist in VBF, which

negatively affects packet delivery ratio. Figure (2.8) demonstrates the

influence of fixed routing pipe on VBF. Packets from nodes A and C are

unable to reach the sink because no nodes exist in the pipe.

• Highly sensitive “routing pipe” radius threshold.

HH-VBF Protocol Overview

Problems solutions in VBF, offer a protocol, named Hop-by-Hop Vector-

Based Forwarding (HH-VBF).HH-VBF that uses the routing vector notion

similar to the VBF protocol.


38

Figure (2.8): VBF using single routing pipe

But, HH-VBF uses various virtual pipes from forwarders nodes to the sink

instead of use of single vector in VBF. Thus, each node can make the adaptation

for the packet routing decisions depending on its current position. From these

characteristics the following interests are made:-

• No need to increase the radius of pipe after the transmission range to

enhance performance in routing, because of there is routing pipe for

each node.

• HH-VBF Strengthens packet delivery ratio compared with VBF,

because of dispersed networks. While the number of nodes qualified is

small. HH-VBF can find a data delivery path as long as it exists. Also,

HH-VBF is less than VBF in sensitivity to routing pipe radius

threshold.

Routing In HH -VBF Protocol

In HH-VBF [45], the routing pipe is redefined to be a virtual pipe from

source to sink instead of single vector pipe. This policy allows for finding a

routing path in expanded manner when compared with VBF. Suppose a node N

receives a packet from the source or any through node S, after receiving the

packet, the N calculates the vector from the S to the sink. That offers to the

forwarding pipe changing each hop in the network. For this reason it is named

hop-by-hop vector based forwarding (HH-VBF). Upon a reception it calculates

the vector from its sender to the sink, and computes its distance to that vector. If

the distance is less than the predefined threshold, it will drop the packet. Time

period is represents the time a node carries the packet before forwarding it. The

self-adaption algorithm in HH-VBF is different from that in the original VBF.

Due to effective suppression strategy package approved in VBF, it can select


39

just a few paths to route packets. That can cause problems in dispersed

networks. To improve the delivery ratio in dispersed networks, some repetition

control is introduced in the self-adaption procedure for HH-VBF. In HH-VBF

[38] when a node receives a packet, it first carries the packet for certain time

period suit with its desirableness factor (the same VBF). Therefore, nodes have

small desirableness factor that will be sent first. In this manner, each node in the

neighborhood can hear the same packet several times. HH-VBF allows each

node to calculate its distances to the different vectors from the forwarding

packets to the sink. If the smallest distances of them stay greater than smallest

distance threshold already calculated, this node will forward the packet; else, it

drops the packet. Clearly, the largest threshold will permit to forward packets.

So, HH-VBF can control forwarding redundancy by tuning that smallest

distance threshold. Each node uses the self-adaptation algorithm to reduce

excessive packets. Figure (2.9) shows a high quality concept of HH-VBF using

the same network setting as VBF. As shown, in HH-VBF, nodes A and C can

access to the sink by using way that are not allowed with VBF [46].

Figure (2.9): HH-VBF with per-hop vector computing.

2.6.2.3 Vector-Based Void Avoidance (VBVA )


40

VBVA is expansion of VBF routing protocol. In the absence of void, the

routing in VBVA is the same as routing in VBF. If a void existed, VBVA use

one of the two mechanisms that have vector-shift mechanism or back-pressure

mechanism to overcome the void.

Void Detection

A node detects the existence of a void by hearing packet transmission of the

adjacent nodes [47]. The starting point and the ending point of the vector are

given the symbols S and T respectively. In any node N, the forwarding vector of

the packet is the projection of the vector SN on the forwarding vector ST. A

node is a void node if all the advances of its neighbors on the forwarding vector

carried in a packet are smaller than its own advance. As shown in Figure (2.10),

the forwarding vector of a packet is ST, and the advances of nodes B, C and F

on the forwarding vector are named as AB, AC and AF, respectively. All the

neighbours of node F have smaller advances than F on the forwarding vector ST.

Therefor node F is a void node.

Figure (2.10): An example of void node.

VBVA mechanisms

There are two mechanisms used in VBVA for overcoming the void:


41

1. Vector-Shift Mechanism

When a node identifies the void node for a packet, it will try to avoid the

void by shifting the forwarding vector of the packet. To accomplish this, it will

send a broadcast message containing a vector-shift packet by node to all its

neighbor nodes. Upon receiving this message; each node outside the forwarding

pipe will try to forward its data packet to a new vector from them to the sink.

Figure (2.11 (a)) shows this mechanism [47].

2. Back-Pressure Mechanism

A node that found to be end node will send a controlling packet, called

BP (Back Pressure) packet. Upon receiving a BP packet, every neighboring

Node tries to shift the forwarding vector of the corresponding packet if it has

never shifted the forwarding vector of this packet before. Otherwise, the node

broadcasts the BP packet again. The BP packet routed in reverse side move

from the sink until it reaches a source node that can act on the vector shifting

it to pass the packet towards sink. Figure (2.11 (b)) shows an example for the

back-pressure process.


42

(a) Vector shift mechanism (b) Back pressure mechanism

Figure (2.11): Two mechanisms of VBVA

2.7

Water quality assessment

2.7.1 Background

The rate of increase in urban, agricultural, and industrial activities has

raised scientists’ concerns about environmental issues and in particular about

water pollution. Therefore, much effort has been made in the development of a


43

comprehensive index that is representative of the overall water quality. Several

authors have integrated water quality variables into indices, technically called

water quality indices (WQIs) ([48], [49], [50] and [51]). Most are based in a

concept developed by the U.S. National Sanitation Foundation [52].

Since, several modified water quality indices (WQI) have been developed

based on the WQI. The WQI is obtained by adding the multiplication of the

respective weight factor by an appropriated quality-value for each parameter.

The WQI index consists of nine parameters: dissolved oxygen (0.17), fecal

coliforms (0.16), biochemical oxygen demand (0.11), pH (0.11), temperature

change (0.1), phosphates (0.10), nitrates (0.10), turbidity (0.08) and total solids

(0.07). In parentheses are given the weight factors according to the importance

of the parameters [8].

However, use of these traditional indices raises many problems. One of the

most important problems is that values with different distances from a limit have

the same effect on the final index score [9]. This leads to an unclear distinction

between each mode of the index and causes inaccuracies and ambiguity when

making decisions about boundary values [8]. The most critical deficiency of

these indexes is the lack of dealing with uncertainty and subjectivity present in

this complex environmental problem [10].

In recent years, artificial intelligence (AI) computational methods such as

knowledge-based systems, neural networks, genetic algorithms, and fuzzy logic,

have been increasingly applied to environmental issues [53]. Fuzzy logic was

introduced by Zadeh (1965) and has become one of the most common

approaches in the field of AI. It is believed to be appropriate for developing

environmental indices, because it has the ability to reflect human thoughts and

expertise in the indices, enabling them to deal with non-linear, uncertain,

ambiguous and subjective information. It also enables us to include parameters


44

with different values and meanings in the index, including both qualitative and

quantitative variables. Furthermore, it is a reliable method for reporting the

results of an assessment in linguistic terms, which are understandable for the

public, managers, decision-makers, and non-experts in general [54].

The use of linguistic variables to describe and assess complex systems has

already been extensively elaborated by computer scientists, the fuzzy logic. One

of the main advantages of fuzzy logic is the ability to model expert human

knowledge, a necessary feature to be considered in the complex process of

environmental management.

The term fuzzy inference system embraces a wide set of diverse

methodologies intended to deal with uncertainty and subjectivity. Since its

introduction in 1965 by Lofti Zadeh, fuzzy inference system has been applied to

many research areas. The interest in fuzzy is still growing through years from

2002 to 2011. The number of papers related to “fuzzy inference system” for

2002 to 2011was 1700 papers while the two terms "fuzzy" and "water quality"

together was only 98 papers. This indicates that it used only 5.7 % of the total

1700 in the field of water quality. Further scientific advances within this field

and their widespread acceptance and use are expected to follow.

2.7.2 Implementing Materials and methods

2.7.2.1 Fuzzy inference system

Fuzzy set theory has been developed for modeling complex systems in

uncertain and imprecise environment [55]. Fuzzy logic uses sets with unclear


45

boundaries. Fuzzy logic can be used for mapping inputs to appropriate outputs.

Figure (2.14) shows an input–output map for the water quality classification

problem: “Given a comprehensive set of water quality indicators, what is the

water condition in a river?” Water quality indicators and river condition are

fuzzy definitions, since they do not present clearly defined boundaries.

Figure (2.12): Input–output map for the river water quality problem in a fuzzy

inference system.

Fuzzy inference is the process of formulating the mapping from a given

input to an output using fuzzy logic. Developing an index based on the fuzzy

logic necessitates comprehension of three important parts of the fuzzy inference

system, including membership functions, fuzzy set operations and inference

rules, which are briefly described below. Each selected input or input set has a

domain called the universe of discourse that is divided into subsets which are

Water

quality

indicator 1

Water

quality

indicator 2

Water

quality

indicator n

∑

Water status:

Excellent

Good

Average

Poor

Verypoor

IF-Then Rules

(Mamdani)

“All rules are evaluated in parallel using fuzzy reasoning”

" Inputs are crisp numbers limited to a range"

“Results of rules are aggregated”

“The result is a crisp number”


46

expressed by linguistic terms. The relationships between the subsets of inputs

and outputs, as well as those among the subsets of inputs, are defined by if-then

rules and fuzzy set operators.

2.7.2.2 Membership functions

A membership function is a curve that defines how each point in the input

space is mapped to a membership value between 0 and 1. The input space is

called the universe of discourse. The output-axis is called the membership value

μ. If X is the universe of discourse and its elements are denoted by x, then a

fuzzy set A is defined as a set of ordered pairs.

A = {x, µ A (x) | x ∈ X} (2.2)

Where μA(x) is the membership function of x in A. A membership function is

an arbitrary curve whose shape is defined by convenience [55].

2.7.2.3 Fuzzy set operations

The standard fuzzy set operations are: union (OR), intersection (AND) and

additive complement (NOT). They manage the essence of fuzzy logic. If two

fuzzy sets A and B are defined on the universe X, for a given element x belong to

X, the following operations can be carried out [55]:

Intersection, AND: µ A ∩ B (x) =min (µ A (x), µ B(x)) (2.3)

Union, OR: µ A ∪ B (x) =max (µ A (x), µ B(x)) (2.4)

Additive complement, NOT: µA-(x) = 1- µΑ(x) (2.5)


47

2.7.2.4 Inference rules

Inference rules define the relationships among the subsets of the inputs and

outputs. This process is carried out by if-then rules which generate a new output

subset. Each rule consists of two parts including an “if-” part, which is called the

antecedent, and a “then-” part, which is called the consequent. The if-then rules

are written in the following way:

IF A is a THEN C is c.

IF B is b THEN C is c.

Where a, b, and c are the linguistic terms for the subsets defined for sets A,

B, and C, respectively [55].

Chapter

Three

UWSNs Simulation

and Network

Results

Chapter Three….......................................UWSNs Simulation and Network Results

47

Chapter Three

UWSNs Simulation and Network Results

3.1 Introduction

This chapter is divided into two parts, select the appropriate simulator and

the other is the implementation and assessment of the network. A comparison is

made among the most important existing simulators and the motivations for

choosing aqua-sim simulator from all existing simulators. Then evaluating of the

performance of several protocols is made through the implementation of

different scenarios which were compared according to the system requirements.

Finally the last scenario for pollution monitoring system is implemented and the

results obtained are analyzed to assess performance.

The structure of the proposed system to monitor pollution in the rivers of

Iraq and main stations is shown in Figure (3.1). After completing the

implementation of this system should produce some clear growth of monitoring

systems, although, started from the traditional way of pollution monitoring as

described in Figure (3.2). However, the researcher is fond of this field and will

continue research in its aspects.

3.2 UWSN simulation and Simulation Design

In order to help simplifying the study in underwater sensor networks, the

researcher found it appropriate to have a regular simulation platform for

matching and assessing different network designs, algorithms and protocols.

In the literature, there are certain efforts to integrate audio broadcast models into

the simulation of underwater audio networks.


48

Figure (3.1): Main Stations of a Pollution Monitoring System

Figure (3.2): Evolution of Monitoring Systems.

Underwater Wireless Sensor Network

Final Report

Knowledge

Sensor Data

MATLAB

(FIS)

Physical Value

Human

Physical Value

Sensor Human

Physical Value

Sensor Wireless Human

Physical Value

Human Data Assessm

Wireless

Sensor


49

The researcher found that there is no broad packet level underwater sensor

network simulator that is available up till now. There are several packet level

network simulators that are used broadly such as NS-2 [56] and OPNET [57].

However, they were developed for terrestrial radio wireless and/or wired

networks, not for underwater sensor networks. They cannot be used for the

simulation of underwater sensor networks without important changes for the

causes stated in chapter one.

A simulator, called Aqua-Sim, was developed for underwater sensor

networks by [58]. NS-2 is selected as the progress platform since NS-2 is a very

powerful, widely used, and open source simulator. NS-2 provides effective and

suitable ways to arrange network and nodes. NS-2 has been commonly used in

sensor network simulations [59]. Use of NS-2 is that it is open source, famous

and generally used by researchers. This means that anybody willing to fulfill its

explanation for executing NS-2 simulations can use the same code [60].

Notwithstanding a hard tuning of the broadcast model parameters, NS-2 offers

the results closest to authenticity in the indoor environment [61]. Two languages

are used in NS-2, C++ and Otcl. Users can use Otcl language to simply adjust

the restrictions of protocols and algorithms executed in C++. NS-2 also allows

having the benefit of the abundant existing source codes. NS-2 is broadly used

by researchers; it has rich wireless protocols and services that are possibly

valuable for underwater sensor networks. Most significantly, NS-2 is open-

source, which permits public access and fast advances. As Aqua-Sim is

developed based on NS-2, it can efficiently simulate audio signal reduction and

packet impacts in underwater sensor networks. Furthermore, Aqua-Sim supports

three-dimensional placement and can simply be integrated with the present

codes in NS-2.

Rather than fixing the current simulation package of wireless networks, a

new simulation package, called Aqua-Sim is developed for underwater sensor


50

networks. In NS-2, Aqua-Sim is in parallel with the Carnegie Mellon University

(CMU) wireless simulation package (wireless model allows simulation of

wireless LANs and ad-hoc networks). Figure (3.1) shows the connection

between Aqua-Sim, CMU wireless package, and NS-2 basics.

Aqua-Sim is free of the wireless simulation package and is not affected by

any alteration in the wireless package. In contrast, any alteration to Aqua-Sim is

also restricted and does not have any influence on other packages in NS-2. In

this way, Aqua-Sim can grow freely.

Figure (3.3): Relationship between Aqua-Sim and other packages of NS-2.

Generally, Aqua-Sim has three classes:

1. Network Entity Classes: These types of classes represent concrete network

entities. Physical layer is represented as an “UnderwaterPhy” object and its

broadcast MAC protocol is represented by a “Broadcast- MAC” object. These

objects are required to fulfill their own functionalities and provide standard

interfaces to the upper/lower layer network entities.

2. Pure Interface Classes: These kinds of classes are really simulated and can

never be instantiated. Yet, they identify known interfaces and act as the basis

classes for others. For example, the “UnderwaterMAC” class in Aqua-Sim

offers a known interface to the MAC layer entities such as “BroadcastMAC”

and “RMAC”. Even though the execution of “BroadcastMAC” is rather

dissimilar to that of”RMAC”, they share the same interface to the physical


51

layer and the logic link layer, which is indicated by their basis class

“UnderwaterMAC”.

3. Shared Function Classes: This kind of classes offers some shared jobs to other

classes and can be involved into any classes in Auqa-Sim. Although these

classes do not have instantiations in the consistent network node, they are

rather significant to Aqua-Sim. For example, “UW-hash-table” fulfills a

highly effective hash table, which may be used by the routing and the MAC

protocols. Figure (3.2) schemes the class diagram of an Aqua-Sim.

Figure (3.4): Class diagram of Aqua-Sim


52

3.2.1 Fidelity and flexibility testing of Aqua-sim

In order to test the fidelity of Aqua-Sim, we design experiments to compare

simulation results with real test bed results. The actual tests are achieved in

Aqua-Lab, a lab test bed developed by the Underwater Sensor Network Lab

[62].Figure (3.5) shows the network topology for both simulation and real test.

Figure (3.5): Network topology for fidelity testing

Node 6 is configured as the receiver and all other nodes send data to node 6.

This gives a one hop, multi-source and single-sink topology. Micro-modem [63]

used in Aqua-Lab provides a data rate at 80bps which is used also in the

simulations.

Aloha protocol was implemented for both Aqua-Lab and Aqua-Sim. The

packet length and the ACK message are set to 32 bytes. The maximal number of

retransmissions is 3 and the input traffic of every node follows an exponential

distribution. Two experiments are conducted to compare the simulations and

real tests. In the first experiment, the average input traffic of every node is set to

0.02 packets per second and change the number of data senders from 1 to 5. In

the second experiment, the total average input traffic is 0.1 packets per second

and change the number of data senders (n). Thus, the input traffic of every node

is 0.1 n. The results are plotted in Figure (3.6) and Figure (3.7). These figures

show clearly that Aqua-Sim can closely approximate the results of the real tests


53

under different input traffic patterns. In other words, it can be claimed that

Aqua-Sim could reproduce the real world with high fidelity.

Figure (3.6): Throughput with fixe input traffic per node

Figure (3.7): Throughput with fixed total input traffic


54

3.3 Performance Assessment of UWSN Protocols

This section presents the assessment of the performance of each of MAC and

routing protocols depending on different metrics to choose the most suitable

protocols appropriate for this work.

3.3.1 UWSN MAC Layer Protocols

This section, presents the assessment contention based protocols like

Broadcast MAC, Aloha, R-MAC, FAMA, and UWAN_MAC by comparing

them in terms of energy consume and received throughput. Three of them will

be elected based on previous assessment and compared in terms of total drop

packets and average end to end delay to choose useful MAC protocols for such

topologies.

3.3.1.1 Assesment and Simulation Results

The performance of Broadcast MAC, Aloha, R-MAC, FAMA, and

UWAN_MAC protocols are to be evaluated. All simulations are performed

using the Network Simulator (NS-2) [64], with an underwater sensor network

simulation package (called Aqua-Sim) extension [65]. Table (3.1) lists the

parameters in all simulation scenarios.

The scenario of figure (3.8) is considered. in this scenario firstly the

throughput and energy consumption for the five protocols were considered. In

the simulation, number of nodes are 7, which 6 nodes are sender and one node

(sink) is receiver. The position of each node is chosen as in figure (3.6). Traffic

is generated according to a Poisson process. Each sender node sends packets per

10 second, which can help to effectively avoid the interference between two

continuous data packets.


55

Table (3.1): Simulation Parameters of MAC Protocols Test

Topology area 300 m * 300 m

Topology depth 10 m

Transmission power 0.6 watt

Receive power 0.3 watt

Idle power 0.001 watt

Maximum transmission range 50 m

Routing protocol VBF(vector based forwarding)

Bandwidth Bandwidth 10 Kbps

Frequency 25 kHz

No. of node 7

Channel Underwater channel

Propagation Underwater propagation

Antenna Omni-directional

As shown in Figure (3.9), the FAMA protocol is higher than other protocols

in energy consumption because it does not consider sleep mode taken into

consideration by others. On the other side, R-MAC is the lowest energy

consumption but have the serious drawback is no technique can be proposed for

the node which tends to change its transmission schedule, or when a node fails

or a new node joins the network. Figure (3.10) shows the part of received

throughput and found that the higher three protocols are UW-MAC, Broadcast

and FAMA protocols.


56

Figure (3.8): Simulation Scenario of MAC Protocol Test

To clarify the issue more about the most appropriate protocol in terms of

energy efficiency as well as the rest of the performance specifications a

comparison between the top three protocols is made .They are compare in terms

of average end to end delay (represents the average time taken by a packet to

travel from the source node to any of the sinks) and total drop packets for

different times and the same scenario as shown in Figure (3.11) and Figure

(3.12) respectively. It is found that the delay in the FAMA is irregular and

higher than UW_MAC and Broadcast, either in terms of total drop packets and

found that Broadcast protocol is higher than UW-MAC and FAMA protocol.

Notice that, all simulation results are shown in table (3.2) of appendix A.


57

Figure (3.9): Energy Consumption of UWSN MAC Protocols

Figure (3.10): Received Throughput of UWSN MAC Protocols


58

Figure (3.11): Average End To End Delay of UWSN MAC Protocols

Figure (3.12): Total Drop Packets of UWSN MAC Protocols


59

3.3.2 UWSN Routing Protocols

In this section, a comparison between three routing protocols (VBF, HH-VBF

and VBVA) is presented in term of energy consumption, average end-to-end delay

and packet delivery ratio. Comparison is carried out by using Aqua-Sim simulators

for underwater sensor networks and NS-2 based simulator installed in Linux

environment.

3.3.2.1 Assesment and Simulation Results

In this section, the performance of VBF, HH_VBF and VBVA routing

protocols are evaluated using table (3.3) as parameters in all simulation scenarios.

The broadcast MAC protocol is used, which is suitable for geographic routing

protocols especially VBF [58]. In Broadcast MAC protocol, a node that has

packets to send senses the channel at first. If empty, it broadcasts the packets. Else,

it backs off and dropping the packets if the node backs off 4 times. The source

location is (490, 490, 500) close to the one corner, while the sink location is (10,

10, 0) close to opposite corner at the surface. All other nodes are mobile between

them. Each node randomly chooses destination and move towards that target. As

soon as it reaches the target node, it randomly chooses a new target and it moves in

a new direction.

Three metrics are proposed:

a) Packet Delivery Ratio: - the ratio of the number of success packets that

received at the sinks to the total number of generated packets at the source

node.

b) Average End-to-end Delay: - defined as the average time taken by a packet

for traveling from the source node to the sinks node.

c) Total Energy Consumption: - defined as the total energy consumption in

delivering packet, including transmitting, receiving, and idling energy

consumption of all nodes in the network.


60

Table (3.3) Simulation Parameters of Routing Protocols Test


Topology depth 500 m

Transmission power 2 watt


Idle power 0.001 watt


MAC protocol Broadcast

Bandwidth Bandwidth 10 Kbps

Frequency 25 kHz

Channel Underwater channel

Propagation Underwater propagation


Maximal speed 5 m/s

Minimal speed 0 m/s

Packet size 50 bytes

Simulation time 500 second


61

Simulation results are presented in table (3.4) of appendix A, and in

figure (3.13). The VBF protocol can achieve highest delivery ratio in a

number of nodes less than number of nodes in the HH_VBF and VBVA

protocols. Therefore, nodes that achieve 100% packet delivery ratio are

considered the optimum number of nodes to obtain high success rate. In

figure (3.14) when compared between protocols in term of e2e delay, found

the VBVA protocol the highest one because of the effort in the mechanical

to prevent voids. In figure (3.15) the important parameters (energy

consumption) are compared. It is seen that VBVA protocol is the lowest

energy consume because of less number of forwarder used in this protocol

which leads to reduce the transmitted power and subsequently reduced total

energy consumption. Depending on the type of system used in pollution

monitoring and to assess any protocol to be appropriate for this work, these

data to be monitored should have high sensitivity and importance, number of

nodes required, the delay should not affect the monitoring process, cancel of

some packets to affect the validity of the information and other things that

can deduced and find solutions to the problems of these systems.

Figure (3.13): Packet Delivery Ratio of UWSN Routing Protocols


62

Figure (3.14): Average End-To-End Delay of UWSN Routing Protocols.

Figure (3.15): Energy Consumption of UWSN Routing Protocols


63

3.3.3 Final Design

In this section, the final design of UWSN will be assessed for pollution

monitoring system in terms of energy consume, received throughput, total drop

packets and average end to end delay.

3.3.3.1 Assessment and Simulation Results

This section presents the evaluation of the performance of the final design

of UWSN for pollution monitoring system. Figure (3.16) represents the structure

of the network consisting of the two networks are UWSN and the terrestrial

network. UWSN is composed of 7 nodes, 6 of them will be sent and one

represents sink node to collect data or values of pollutants and then passed to the

terrestrial network through the antenna outside of the water surface. The

terrestrial network is represented by the on shore or base station linked by

satellite to the pollution center where data is processed and monitored. Using

table (3.5), as parameters in simulation scenario, the steps to implement the

network in the simulator, as well as full source code are available in Appendix

A.


64

Table (3.5): Final design Simulation Parameters


Topology depth 10 m

Transmission power 0.6 watt


Idle power 0.0 watt


UWSN Routing protocol VBF

Terrestrial net. Routing protocol AODV

UWSN MAC protocol UW-MAC

Terrestrial net. MAC protocol IEEE 802.11

Vector width 100 m

UWSN Bandwidth Bandwidth 10 Kbps

UWSN Frequency 25 kHz

No. of node 7

No. of base station 1

UWSN Channel Underwater channel

UWSN Propagation Underwater propagation



65

Figure (3.16): Network Topology of Final Design

In the simulation, number of nodes are 7, from which 6 nodes are sender

and one node (sink) are receiver and one base station (B.S). Traffic is generated

according to a Poisson process. Each sender node sends one packets per 10

second, which can help to effectively avoid the interference between two

continuous data packets.

According to [58], which state that "the broadcast MAC protocol was

suitable for geographic routing protocols especially VBF". In present work

scenario, and through experience using a number of metrics such as (Received


66

Throughput, Total Drop Packets, Average End to End Delay and Energy

Consumption) proved that UW-MAC protocol is best suited with geographic

routing protocols especially VBF and therefore is best suited among the rest

underwater MAC protocols in accordance with VBF as shown in table (3.6) of

appendix A and Figures (17 – 20).

It is seen from figure (3.20) that using UW-MAC protocol with VBF

protocol achieve to highest received throughput approximately to 6.5Kbp and

whenever the number of nodes is increased the received throughput increase.

Due to the mechanism of work of UW-MAC protocol with VBF protocol the

work together making the system or network less dropped packets as can be

seen in figure (3.17).

Figure (3.18) shows that the network achieve less end to end delay

approximately to 236 MS and in all run periods, compared to the degree of

importance of this information. It is possible to call the system a real-time

system.

Figure (3.17): Total Drop Packets of UWSN in Final Design


67

Figure (3.18): Average End to End Delay of UWSN in Final Design

Figure (3.19): Energy Consumption of UWSN in Final Design


68

Figure (3.20): Received Throughput of UWSN in Final Design

The data (pollutants) has reached from UWSN into the surface base station

(shore) sent to the pollution center by GPRS or Satellite. Figure (3.16) Satellite

is used for several reasons; the most important reason is the permanent presence

service under normal circumstances. In terms of cost it is less expensive and the

data sent does not exceed 100 kbps for short range of less than 1 kilometer [66].

Available bandwidth for different ranges in UW-A channels are shown in the

table (3.7).

All of the above is for one cluster in the case of a number of clusters in

several regions for pollution monitoring in which we will repeat the UWSN

cluster and then linking it with the shore.


69

Table (3.7): Available bandwidth for different ranges in UW-A channels.

Bandwidth (kHz) Range (km)

<1 1000 Very long

2–5 10–100 Long

≈10 1–10 Medium

20–50 0.1–1 Short

>100 <0.1 Very short

The data has been collected reach directly to the data file (online), in the

next chapter presents the details of the traditional way to assess water quality

and compare it with a novel way in Iraq that use fuzzy inference system.

3.4 Summary of contributions and achievements

It can be concluded that the FAMA protocol was not suitable for UAWSN in

terms of energy efficiency. From the other four protocols, R-MAC is more energy

efficient. In terms of throughput, UW_MAC is the best protocols, then Broadcast

protocol and then Aloha protocol. An improvement of 30 % compared with other

protocols was found when using UW_MAC protocol, in terms of energy,

throughput, total drop packets and end to end delay. Hence, UW-MAC protocol is

suggested for the monitoring system in this work.

The conclusions drawn are, if the number of nodes is 50 nodes or less, the

VBF protocol is the most appropriate protocol than other routing protocols. If

the number of nodes is 50 and by using VBF protocol will get a system less

energy consumption, less delay and high packet delivery ratio. Thus, the fewest


70

number of nodes will be used to achieve approximately 100% success rate

thereby obtaining a less expensive system that overcomes the problems of using

redundant node to achieve the same purpose.

But if the data monitored is sensitive in terms of delay then VBVA and HH-

VBF protocols are more suitable than VBF protocol. If the monitoring systems

have a large number of nodes, it is best to use VBVA Protocol to overcome the

problem of charging and recharge for long periods as VBVA Protocol is high

energy-efficient. Thus, for the monitoring system, VBF routing protocol is

suggested.

There is a slight difference in energy consumption compared with using

broadcast protocol with VBF protocol. This difference will decay when the

number of nodes is increased making the system more efficient in terms of

energy consumption. However, if the number of nodes is increased, the system

will bring 100% packet delivery ratio in less number of nodes and therefore the

system becomes more efficient in energy consumption than other systems.

Energy consumption of the pollution monitoring system .

Chapter

Four

Development of a

water quality

Assessment Using

FIS

Chapter Four …………. Development of a water quality Assessment Using FIS

71

Chapter Four

Development of a water quality Assessment Using FIS

4.1 Introduction

The present work aims at developing a proposed water quality index in Iraq

based on fuzzy inference system. That is, comprehensive artificial intelligence

(AI) approaches to the development of environmental indices for routine

assessment of water quality, particularly for human drinking purposes. Multiple

parameters were included based on their critical importance for the overall water

quality and their potential impact on human health. To assess the performance of

the proposed index under actual conditions, a case study was conducted at Tigris

River, Iraq, employing water quality data of various sampling sites in the water.

4.2 Calculations of Traditional WQI in Iraq

The WQI was calculated using the Weighted Arithmetic Index method. The

quality rating scale for each parameter (qi) was calculated by using equation (1)

[51]:

qi = �CiSi� ∗ 100 (4.1)

A quality rating scale (qi) for each parameter is assigned by dividing its

observed concentration (Ci) in each water sample by its respective standard

value (Si) for each sample (i) and the result is multiplied by 100. Relative

weight (wi) was calculated by a value inversely proportional to the

recommended standard value (Si) of the corresponding parameter using equation

(2) [51]: �i = 1�� (4.2)


72

The overall WQI was calculated by aggregating the quality rating scale

(qi) with the unit weight (wi) linearly in equation (3) as follows [51] :

�� = �∑ �i ∗ ��=��=1 � (4.3)

Where: qi: the quality of the ith parameter.

wi: the unit weight of the ith parameter.

n: number of the parameters considered.

Generally, WQI is to be discussed for a specific and intended use of water.

In this study the WQI for drinking purposes is considered an available WQI if its

value is 100 using equation (4) [51]:

Overall WQI = �∑ �i∗��=��=1 �∑ �i�=��=1 (4.4)

The water quality index (WQI) of the Tigris river within Baghdad city has

been calculated using the weighted arithmetic index method by ten parameters

of raw water that were studied in respect to their suitability for human

consumption compared with the standards of drinking water quality

recommended by the World Health Organization [67].

The WQI and overall WQI of all the samples taken were calculated according to

the procedure explained above and the results are presented in Table (4.1).

Based on the WQI value, water is categorized into five groups ranging from

excellent water to water unsuitable for drinking. The computed annual overall

WQI for all samples from 2004 to 2010 was 275.53, which implies that the

water is generally "very poor quality".


73

4.3 Fuzzy Inference Systems, Step by Step

The procedure carried out within a FIS is here described. We have

hypothesized that the levels of dissolved oxygen (DO) and (PH) are sufficient to

evaluate water quality (WQ) by means of an aggregated index called the Fuzzy

Water Quality (FWQ) index. We have chosen “very low”, “low”, “fair ” , “high”,

and “very high” fuzzy sets for inputs, and “very poor”, “poor”, “average” ,

“good” and “excellent” fuzzy sets for the output. Trapezoidal membership

functions define these fuzzy sets Figure (4.1).

Table (4.1): Computed WQI values for Tigris River within Baghdad city from

2004-2010.

Test Type Ci Si wi qi WQI

(wi*qi)

Turbidity 62.18 5 0.2 1243.6 248.72

Alkalinity as 145.16 150 0.01 145.16 1.451

T.Hardness as 328.50 500 0.01 328.5 3.285

pH 7.98 7.5 0.133 106.4 14.15

temp 21.3 19.5 0.05 109.32 5.46

Sulfate as SO4 208.95 40 0.004 83.58 0.334

Total Dissolved 530.26 15 0.002 106.052 0.212

Iron as Fe 1.68 0.3 3.33 560 1864.8

Nitrate as NO3 0.80 50 0.02 1.6 0.032

Ammonia as NH3 0.11 0.2 5 55 275 ∑ �i =�=10�=1

8.759

∑ �i ∗ ��=10�=1 =

2413.444

Overall WQI = �∑ �i∗��=10�=1 �∑ �i�=10�=1 = 275.53


74

Figure (4.1): Membership functions for PH, DO and FWQI parameters.


75

In water quality assessment, the expressions are frequently used by the

experts: “if the levels of PH in a river are low, and the levels or dissolved

oxygen are low, then the expected water quality is poor”. In fuzzy language, it

could be enunciated as figure (4.2).

Figure (4.2): Fuzzy language Rules for the inputs PH and DO.


76

Two river samplings points were used “S1” and “S2”, having PH, DO

values of (1, 10) and (7.43, 5.56) respectively. Before any calculation, an expert

would infer the water quality status in the S1 point by applying the first rule.

However, when input values are close to boundaries between a fuzzy set and

another one, as in S2 point, the output is not so direct, and fuzzy operations

should be carried out.

We must fuzzify the inputs for S2 point. According to membership functions

in Figure (4.1), a value of 7.43 for PH belongs to “fair” and “high” fuzzy sets

are found. Similarly, a value of 5.56 for DO belongs to “low” and “fair” fuzzy

sets. Thus, a variable could belong to more than one set.

The degree of support for the entire rule is used to shape the output fuzzy set.

The consequent of a fuzzy rule assigns an entire fuzzy set to the output. This

fuzzy set is represented by a membership function that is chosen to indicate the

qualities of the consequent. If the antecedent is only partially true, having a

value lower than 1, the output fuzzy set is truncated at this value. This procedure

is called the minimum implication method [55]. This is shown in Figure (4.3),

where columns refer to the input/output fuzzy sets, and rows are the fuzzy rules.

Since decisions are based on the testing of all the rules in the system, these must

be aggregated to make a decision. As depicted in Figure (4.3), output fuzzy sets

for each rule are aggregated to a single output fuzzy set. The aggregation

procedure used here is the maximum method [55], which is the union of all

truncated output fuzzy sets.

The final step is the defuzzification. The input for the defuzzification process is

the aggregated output fuzzy set.


77

Figure (4.3): Fuzzy inference diagram for the water quality scoring problem

with two variables and five rules.

As much as fuzziness helps the rule evaluation during intermediate steps,

the final desired output is a numeric score. The defuzzification method preferred

is the centroid, which is the most prevalent and physically appealing of all

available methods [55].

Finally, Figure (4.4) illustrates the relationships between two of the

parameters included in the index (e.g. DO and PH) and their effect on the final

index score. All computations were carried out using the “fuzzy logic toolbox”

in MATLAB version 7.6.0 (R2008a).


78

Figure (4.4): A surface graph representing the interactions between DO, PH and

the final index value.

The above describes the procedure used to deal with information in a FIS.

In Section 4.4, we describe the use of a fuzzy inference system to classify water

quality in the Tigris River as example. A comprehensive set of 8 water quality

indicators and 40 rules has been used.

4.4 Development of the Fuzzy Water Quality index (FWQI).

A fuzzy index has been developed for the water quality assessment. Ranges

and weights of the variables in the inference system have been taken from the

standard specification of drinking water in Iraq [68], to calculate the quality of

water in traditional way (WQI), as well as using fuzzy inference system

(FWQI).


79

Ten variables taken for the years from 2004 to 2010 for the eight sites in

the Tigris River to calculate the overall WQI in the traditional way, as well as

using FWQI.

To calculate the FWQI eight variables were taken from five sites in three

months, the indicators are the following:

Dissolved oxygen (DO), pH, Turbidity, total Hardness, Iron as Fe, Temperature,

Mn and Cu.

Trapezoidal membership functions were used to represent “very low”, “low”,

“Fair”, “ high” and “very high” for input and “very poor”, “poor”, “average ”,

“good” and “excellent” for output fuzzy sets.

The membership function parameters are a, b, c, and d. A list of these

parameters is summarized in Table. Ranges for fuzzy sets were based on the

Standard Specification of drinking water in Iraq and guidelines for drinking-

water quality [68] and [67] respectively. These ranges are also shown in Table

(4.2).

4.5 Results and discussion of FWQI

The water condition for the Tigris River has been assessed with the FWQ

index. Input data extracted from public databases have been used to assess water

quality between 2004 and 2010 to assess the performance of these years, and to

compare between traditional WQI and present method FWQI, then to compare

between them in multiple sites through three varying months and different in

terms of geographical climate [69].

The results for the global FWQ index calculated according to the FIS and

traditional WQI from 2004 to 2010 are shown in Figure (4.5). A comparison

between the proposed FWQ index and the reputed WQI index through three

months (June, august 2007 and January 2008) is shown in Figures (4.6, 4.7 and

4.8). The FWQI obtained from section 4.2 for calculating the quality of water

for two indicators (PH and DO) is also observed that the result of FWQI is 62.9.


80

Table (4.2): Parameters for membership functions used in the fuzzy inference

system

Indicator Very Low

Low Fair High Very High Ranges

a=b

c d a b c d a b c d a b c d a b c=d

Temp. 0 1.12

5 10.1

3

1.125

10.13

12.38

21.38

12.38

21.38

23.63

32.63

23.63

32.63

34.88

43.88

34.88

43.88

45 0 – 45

PH 0 0.35 3.15 0.35 3.15 3.85 6.65 3.85 6.65 7.35 10.1 7.35 10.1 10.85

13.65

10.85

13.65

14 0 – 14

T.hardness 0 25 225 25 225 275 475 275 475 525 725 525 725 775 975 775 975 1000

0 -1000

DO 0 0.3 2.7 0.3 2.7 3.3 5.7 3.3 5.7 6.3 8.7 6.3 8.7 9.3 11.7 9.3 11.7 12 0 - 12

Fe 0 0.007

0.06 0.007

0.06 0.08 0.1 0.08 0.1 0.15 0.2 0.15 0.2 0.23

0.29 0.23 0.29 0.3 0 – 0.3

Mn 0 0.002

0.02 0.002

0.02 0.027

0.04 0.027

0.04 0.05 0.07 0.05 0.07 0.077

0.09 0.077

0.09 0.1 0 – 0.1

Cu 0 0.02 0.2 0.02 0.2 0.27 0.4 0.27 0.4 0.5 0.7 0.5 0.7 0.775

0.975

0.775

0.975

1 0 – 1

Tur. 0 0.125

1.126

0.125

1.126

1.376

2.375

1.3 2.3 2.6 3.6 2.6 3.6 3.8 4.8 3.8 4.8 5 0 – 5

FWQI Very Poor Poor Average Good Excellent 0 - 100


81

This indicates that the quality of the water is between "average" and "

good". While in the traditional method is 78.28 which indicates that the quality

of the water is "good". This demonstrates the existence of a difference in the

method of assessing the quality of water will therefore be a difference when

processing that water and also a difference in making the right decision for

refinement to be drinkable.

It has been found that the achieved results from FWQI are less than the results

obtained from the regular method by approximately 41.64 % for a period

observed in figures (4.5, 4.6 and 4.7).This percent come from the total average

of these period.

Results obtained from this index needs further discussion if there was

professional team of environmental engineering to know the levels of pollution,

as well as evaluating ways to put its treatments and also use a variance called

(ANOVA) and this is not our specialty.

Figure (4.5): Results for water quality indexes for Tigris River within Baghdad city from 2004-2010


82

0

200

400

600

800

1000

1200

1400

1 2 3 4 5

Ind

ex

Va

lue

Sampling Site

June 2007

FWQI

WQI

0

200

400

600

800

1000

1200

1400

1600

1800

1 2 3 4 5

Ind

ex

Va

lue

Sampling Site

August 2007

FWQI

WQI

Figure (4.6): Comparative between indexes in June 2007

Figure (4.7): Comparative between indexes in August 2007


83

Figure (4.8): Comparative between indexes in January 2008

4.6 Validity of FWQI

Despite the difficulty of verifying of the proposed index, but it is possible

through approximately three points to demonstrate the validity of this index.

Two of them observed through working in FIS and the third is through a

conducted by some researchers in this field. Some of the things relate to the

competence of the environment engineering and possible yield from this study

in the field of water quality assessment and that it's up to the professional's

people to decide.

One of the most important approaches is the systematic method used in the

development of the index, that is, fuzzy inference system. This systematic

method has proved to be suitable for the development of environmental indices,

because it has the ability to reflect the ideas and human technical expertise,

which enables dealing with non-linear information, uncertainty, ambiguous, and

0

5000

10000

15000

20000

25000

1 2 3 4 5

Ind

ex

Va

lue

Sampling Site

January 2008

FWQI

WQI


84

subjectivity. Additionally, the index is based on the linguistic terminology and

results becoming more understandable for the public, managers and non-experts

peoples. The second approach addresses the inclusion of the parameters in the

index. The presented index can include all of the parameters that were

considerable importance for the overall quality of drinking water and can greatly

affect to health of human. And this is an indication of the necessity for

containing all of these parameters should be included in the index if it is to be

representative of the overall water quality.

Finally, according to Ref. [53], the robustness of the index is

approximately similar to the index that has been presented by a sensitivity

analysis. For this purpose the final rule set of the index is considered as the

baseline, and then measurement of the changes/deviations, were intentionally

made, from the baseline as the percentage of change/variation in the rule sets of

different stages. In fact, the presence of change in the rules of each stage is

obtained by calculating the number of rules that differ from that of the baseline

rule set. For instance, considering a total of 25 rules in the rule set of the final

stage, if 5 rules had been manipulated in the rule set of this stage, the percent of

change/variation would have been 20%. This suggests that the changes in the

rules of different stages of the fuzzy-based index do not significantly change the

index outputs. Implying that the index is quite robust against changes.

Chapter Five

Conclusion and

Suggestions for

Future Work

Chapter Five ……………………. Conclusion and Suggestions for Future Work

85

Chapter Five

Conclusion and Suggestions for Future Work

5.1 Conclusion

The proposition of a system for pollution monitoring in water by using

underwater wireless sensor networks presented. Then, design a methodology to

assess the data collected from this network. The work is divided into two parts;

firstly performance assessment of UWSN through comparing multiple MAC and

routing protocols using specific metrics as energy consumption, received

throughput, total drop packets, average end to end delay and Packet delivery ratio

then elected the appropriate one based on system requirements. The second part

presents a novel way in Iraq that use fuzzy inference system to assess water quality

on data collected from our UWSN.

The conclusions are drawn from the obtained results of simulation that deals

with UWSN as follows:

1. The simulation Aqua-Sim can efficiently simulate audio signal reduction

and packet impacts in underwater sensor networks. Also, Aqua-Sim can

simply be integrated with the present codes in NS-2.

2. The FAMA protocol is not suitable for UAWSN in terms of energy

efficiency. The rest four protocols, R-MAC protocol is more energy

efficient. In terms of throughput, UW_MAC is the best protocols, then

Broadcast protocol then Aloha protocol. Thus, improvement of 30 %

compared with other protocols has been found when using UW_MAC

protocol, in terms of energy, throughput, total drop packets and end to end

delay.

3. If the number of nodes is 50 nodes or less, the VBF protocol is the most

appropriate one. If the number of nodes is 50 then using VBF protocol get a


86

system less energy consumption, less delay and high packet delivery ratio.

Thus, the fewest number of nodes will be used to achieve approximately

100% success rate. A less expensive system that overcomes the problem of

use redundant node to achieve the same purpose is obtained.

4. Due to the mechanism of work of UW-MAC protocol with VBF protocol the

work together making the system or network less dropped packets. Also, it

achieved less end to end delay approximately to 236 MS and in all run

periods, which can be assumed as a real-time system.

5. Broadcast protocol with VBF protocol is slightly less than using UW_MAC

protocol with VBF protocol. This difference can decay when comparing

highly success rate has been obtained from final design.

The conclusions observed from using fuzzy inference system to develop a water

quality index are as follows:

1. A robust decision making tool has been presented for water assessment and

management in the form of an index called fuzzy water quality index

(FWQI).

2. Fuzzy inference system that has flexibility in its work to develop

classification models should be recommended in the development of similar

environmental indexes.

3. FIS can put a baseline index to be used with other parameters with simple

changes thereby do not change the score of index significantly, but in the

same number of parameters of baseline.

4. FIS is a reliable method for reporting the results of the monitoring system in

linguistic terms, which are understandable for the public and non-experts

people to decide.


87

5.2 Suggestions for Future Work

The monitoring system has been proposed has many points as suggestion

for future work that makes a pollution monitoring systems more efficient. These

points are summarized as follows:

1. The information needs to transmit is decided according to water quality

grade which is quantitatively analyzed directly by using fuzzy

comprehensive evaluation rather than transmitting large amount of raw

data to the monitoring center therapy solve the problem of high energy

consumption.

2. Plan to manipulate the VBVA protocol to make it higher packet delivery

ratio and less end to end delay to achieve more energy efficient and high

data rate routing protocol at the same time.

3. Development of a DBR routing protocol in Aqua-sim simulator by using

the network programing tools represented by using TCL language and

C++.

4. Formation of a working group with the engineering environment or the

number of specialists in the field of pollution to set effective limits for

indicators and consistent with the Iraq's water environment and therapy.

There will be a high level of water quality assessment using fuzzy,

because the degree of knowledge will be increased.

5. Design of different GUI to add on pollution monitoring system by using

GUI in MATLAB to alert when the indicator limits exceed.

88

REFRENCES

[1] X. (Sherman) Shen, Y. Pan, "Fundamentals of Wireless Sensor

Networks", Book chapter, John Wiley & Sons Ltd, 2010.

[2] I. F. Akyildiz, M. Can ,"Wireless Sensor Networks ", Book chapter,

John Wiley & Sons Ltd.2010.

[3] B. Kishore,"Muti-Metric Adaptive Routing Algorithm for Underwater

Wireless Sensor Networks", Master thesis, Texas A&M University, 2011.

[4] D. Pompili, T. Melodia,"An architecture for ocean bottom Under

Water Acoustic Sensor Networks," poster presentation, in Proc. of the

Mediterranean Ad Hoc Networking Workshop (MedHocNet), Bodrum,

Turkey, July 2004

[5] U.Devee Prasan, Dr. S. Murugappan," Underwater Sensor Networks

:Architecture, Research Challenges and Potential Applications ",

International Journal of Engineering Research and Applications, Vol. 2,

Issue 2,Mar-Apr 2012.

[6] Jun-Hong Cui, Jiejun Kong, Mario Gerla, Shengli Zhou," Challenges:

Building Scalable Mobile Underwater Wireless Sensor Networks for

Aquatic Applications", IEEE Network, Special Issue on Wireless

Sensor Networking, Vol. 20, No. 3, pp. 12-18, May/June 2006.

89

[7] M.P. Malumbres, P. Pablo Garrido, Carlos T. Calafate,Jose Oliver,"

Underwater Wireless Networking Technologies ", Progress In

Electromagnetics Research, Vol. 95, pages 219-239, 2008.

[8] Andre Lermontov, Lıdia Yokoyama , Mihail Lermontov , Maria

Augusta," River quality analysis using fuzzy water quality index: Ribeira

do Iguape river watershed, Brazil", Ecological Indicators, Volume: 9

Issue: 6 Pages: 1188-1197, 2009.

[9] Gabrielle Thie´baut, Guillaume Tixier, Franc¸ois Gue´rold,Serge

Muller," Comparison of different biological indices for the assessment of

river quality:application to the upper river Moselle (France)",

Hydrobiologia, Volume: 570 Issue: 1 Pages: 159-164, Springer, 2006.

[10] William Andrés Ocampo Duque," the development of decision-

making systems based on fuzzy models to assess water quality in rivers",

PhD Thesis, Department of Chemical Engineering Universitat Rovira i

Virgili,Tarragona – 2008.

[11] Feng Zhang," Underwater Sensor Networks for Water Quality

Monitoring" Project Final Report, 2008

[12] Arijit Khan, Lawrence Jenkins," Undersea wireless sensor network

for ocean pollution prevention", International Conference on

Communication Systems Software and Middleware, Pages: 2-8 Provider:

IEEE، Bangalore, 2008.

[13] WANG W., LIN Y.," The Application of Wireless Sensor Network

on Regional Environmental Protection", Conference on Regional

90

Economy and Sustainable Development, ISBN： 978-0-646-50352-3,

Nov. 2008.

[14] Peng J. , Hongbo X. , Zhiye H. and Zheming W.," Design of a Water

Environment Monitoring System Based on Wireless Sensor Networks ",

Sensors , Volume: 9 Issue: 8 Pages: 6411-6434, 2009.

[15] Nader M. , Imad J., Jameela J. and Liren Z.," Sensor Network

Architectures for Monitoring Underwater Pipelines ", Sensors , Volume:

11 Issue: 11 Pages: 10738-10764, 2011.

[16] Chunhu Liu ,Bin Fu ,Han Zhang, Lian Lian." Construction and basic

performance tests of an underwater monitoring network", Artificial Life

and Robotics, Volume: 16, Issue: 1, Pages: 98-101, Provider: Springer,

2011.

[17] Sebastian B. ," Enabling Autonomous Environmental Measurement

Systems with Low-Power Wireless Sensor Networks", Master Thesis,

Mid Sweden University, 2011.

[18] Mustafa S., Mahmood A., Shahzad A. Malik, Izhar-ul-Haq," Design

of Underwater Sensor Networks for Water Quality Monitoring", World

Applied Sciences Journal, Vol. 17 ,page 1441-1444,2012.

[19] Teng Wang," Water Quality Monitoring System based on WSN",

Master Thesis,Royal Institute of Technology,2012.

91

[20] Huma Zia, Nick R. Harris, Geoff V. Merrett , Mark Rivers , Neil

Coles," The impact of agricultural activities on water quality: A case for

collaborative catchment-scale management using integrated wireless

sensor networks", Computers and Electronics in Agriculture , Volume: 96

, Pages: 126-138 Provider: Elsevier,2013.

[21] Ian F. Akyildiz, Dario P., Tommaso M.," Underwater acoustic

sensor networks: research challenges", Ad Hoc Networks, Volume: 3,

Issue: 3, Pages: 257-279, Provider: Elsevier, 2005.

[22] Ian F. Akyildiz, Dario Pompili, Tommaso Melodia," State-of-the-art

in Protocol Research for Underwater Acoustic Sensor Networks", ACM

Mobile Computing and Communication Review by citeseer, 2007.

[23] Xiping Yang, Keat G. Ong, William R. Dreschel, Kefeng Zeng,

Casey S. Mungle, Craig A. Grimes," Design of a Wireless Sensor

Network for Long-term, In-Situ Monitoring of an Aqueous Environment"

, Sensors , Volume: 2, Issue: 11, Pages: 455-472, November 2002.

[24] Lanbo Liu, Shengli Zhou, Jun-Hong Cui," Prospects and Problems

of Wireless Communication for Underwater Sensor Networks ", Wireless

Communications and Mobile Computing, Volume: 8, Issue: 8, Pages:

977-994, Provider: Wiley, 2008.

[25] Bridget Benson," Design of a Low-cost Underwater Acoustic

Modem for Short-Range Sensor Networking Applications", Ph.D. Thesis,

University of California, San Diego, 2010.

92

[26] Daniel L. Codiga, Joseph A. Rice, Paul A. Baxely," Networked

Acoustic Modems for Real-Time Data Delivery from Distributed

Subsurface Instruments in the Coastal Ocean: Initial System

Development and Performance", Oceanic Tech., Vol. 22,page 704-720 ,

2005.

[27] Dario Pompili," Efficient Communication Protocols for Underwater

Acoustic Sensor Networks ", Ph.D. Thesis, Georgia Institute of

Technology, August 2007.

[28] Ian F. Akyildiz, Mehmet Can Vuran," Wireless sensor networks",

Book chapter , A John Wiley and Sons, Ltd, Publication, 2010.

[29]Manjula.R.B, Sunilkumar S. Manvi," Issues in Underwater Acoustic

Sensor Networks", International Journal of Computer and Electrical

Engineering, Vol.3, No.1, February, 2011.

[30] Holger Karl, Andreas Willig, "Protocols and Architectures for

Wireless Sensor Networks", Book chapter, John Wiley & Sons, Ltd.2005.

[31] M.P. Malumbres, P. Pablo Garrido, Carlos T. Calafate, Jose Oliver,"

Underwater Wireless Networking Technologies ", Progress In

Electromagnetics Research, Vol. 95, pages 219-239, 2008, 2008.

[32]Phil Karn," MACA: A New Channel Access Method for Packet

Radio", the proceedings of the 9th ARRL Computer Networking

Conference, London, Ontario, Canada, 1990.

93

[33] Vaduvur B., Alan D., Scott S., Lixia Z.," MACAW: A Media Access

Protocol for Wireless LAN’s", by citeseer Journal, 2009.

[34]Chane L.Fullmer, J.J Garcia-Luna-Aceves, "Floor Acquisition

Multiple Access (FAMA) for Packet Radio Networks", by citeseer

Journal, 1998.

[35] Peng Xie, Jun-Hong Cui," R-MAC: An Energy-Efficient MAC

Protocol for Underwater Sensor Networks", International Conference on

Wireless Algorithms, Systems and Applications , ISBN: 076952981X ,

Pages: 187-198, Provider: IEEE, 2007 .

[36] Asoke K. Talukdar," Mobile Computing ", Book chapter, Chapter

21,page 607, Edition 2, Tata McGraw-Hill Education, Jul 10, 2010.

[37] Farizah Yunus, Sharifah H. S. Ariffin, Yasser Zahedi," A Survey of

Existing Medium Access Control (MAC) for Underwater Wireless

Sensor Network (UWSN)", 2010 Fourth Asia International Conference

on Mathematical/Analytical Modeling and Computer Simulation , Pages:

544-549, Provider: IEEE, 2010.

[38]Min Kyoung Park, Volkan Rodoplu,"UWAN-MAC: An Energy

efficient MAC-Protocol for Underwater Acoustic Wireless Sensor

Networks", IEEE Journal of Oceanic Engineering, Volume: 32, Issue: 3,

Pages: 710-720, Provider: IEEE, July 2007.

[39] Peng Xie, Jun-Hong Cui, Li Lao," VBF: Vector-Based Forwarding

Protocol for Underwater Sensor Networks", In Proc. of IFIP Networking

by citeseer Journal,2005.

94

[40] Charles E. Perkins, Pravin Bhagwat," Highly Dynamic Destination-

Sequenced Distance-Vector Routing (DSDV) for Mobile Computers", by

citeseer Journal, 1994.

[41]P.Jacquet,P.Muhlethaler,T.Clausen,A.Laouiti,A.Qayyum,L.Viennot,"

Optimized Link-State Routing Protocol for AD hoc Networks", Multi

Topic Conference, 2001. IEEE INMIC 2001. Technology for the 21st

Century, Pages: 62-68, Provider: IEEE 2001.

[42] Ian F. Akyildiz, Dario Pompili, Tommaso Melodia," State of the Art

in Protocol Research for Underwater Acoustic Sensor Networks", ACM

Mobile Computing and Communication Review by citeseer Journal,

2007.

[43] Tommaso Melodia, Dario Pompili, Ian E Akyildiz," Optimal Local

Topology Knowledge for Energy Efficient Geographical Routing in

Sensor Networks", IEEE INFOCOM, Volume: 3, Pages: 1705-1716,

Provider: IEEE, 2004.

[44] Peng Xie,Jun-Hong Cui, Li Lao,"VBF: Vector-Based Forwarding

Protocol for Underwater Sensor Networks", In Proc. of IFIP Networking

by citeseer Journal,2009.

[45] Peng Xie, Zhong Zhou, Zheng Peng, Jun-Hong Cui, Zhijie Shi,"

Void Avoidance in Three-Dimensional Mobile Underwater Sensor

Networks", by citeseer Journal, 2010.

95

[46] Nicolas Nicolaou, Andrew See, Peng Xie, Jun-Hong Cui, Dario

Maggiorini," Improving the Robustness of Location-Based Routing for

Underwater Sensor Networks", Conference in OCEANS 2007 - Europe ,

Pages: 1-6, Provider: IEEE, 2007.

[47] Peng Xie, Jun-Hong Cui, Li Lao," VBF: Vector-Based Forwarding

Protocol for Underwater Sensor Networks", International Conference on

Multimedia Information Networking and Security, Pages: 178-181,

Provider: IEEE, February 2006.

[48] Jesus Cabezas Flores," Comments to the use of water quality indices

to verify the impact of C!ordoba City (Argentina) on Suquıa river", Water

Research, Vol.36 , pages 4664–4666, 2002.

[49] Shiow-Mey Liou, Shang-Lien Lo,Shan-Hsien Wang," A Generalized

Water Quality Index For Taiwan", Environmental Monitoring and

Assessment, Volume 96, Issue 1-3, pp 35-52, August 2004

[50]Ahmed Said, David K. Stevens, Gerald Sehlke, "Environmental

Assessment: An Innovative Index for Evaluating Water Quality in

Streams", Environmental Management Vol. 34, No. 3, pp. 406–414,

2004.

[51] Dr. Basim Hussein khudair," Assessment of Water Quality Index

and Water Suitability of the Tigris River for drinking water within

Baghdad City, Iraq", Journal of Engineering, Number 6, Volume 19, June

2013.

96

[52] [NSF] National Sanitation Foundation International, 2007. Available

in: <http://www.nsf.org> (Accessed on October of 2007).

[53] Hamed Gharibi , Amir Hossein Mahvi, Ramin Nabizadeh , Hossein

Arabalibeik ,Masud Yunesian , Mohammad Hossein Sowlat," A novel

approach in water quality assessment based on fuzzy logic", Journal of

Environmental Management , Volume: 112, Pages: 87-95, Provider:

Elsevier ,2012.

[54] Thomas E. McKone, Ashok W. Deshpande," Can Fuzzy Logic Bring

Complex Problems into Focus? Modeling Imprecise Factors in

Environmental Policy", IEEE International Conference on Granular

Computing, Pages: 11-17, Provider: IEEE Aug. 2010.

[55] Timothy J. Ross," Fuzzy Logic with Engineering Applications",

Book chapter , John Wiley & Sons, Ltd,2010.

[56] Geoffrey Xie, John Gibson Leopoldo ,Diaz-Gonzalez,"

Incorporating Realistic Acoustic Propagation Models in Simulation of

Underwater Acoustic Networks:A Statistical Approach", IEEE Journal

of OCEANS , Pages: 1-9, September 2006.

[57] Ethem M. Sozer, Milica Stojanovic, John G. Proakis," Design and

Simulation of an Underwater Acoustic Local Area Network", by citeseer

Journal, 2009.

[58] Peng Xie, Zhong Zhou, Zheng Peng, Hai Yan, Tiansi Hu,Jun-Hong

Cui, Zhijie Shi, Yunsi Fei, Shengli Zhou, "Aqua-Sim: An NS-2 Based

97

Simulator for Underwater Sensor Networks" IEEE Journal of OCEANS,

Pages: 1-7, Oct. 2009.

[59] Yunjiao Xue, Ho Sung Lee, Ming Yang, Priyantha Kumarawadu,

Hamada H. Ghenniwa, Weiming Shen," Performance Evaluation of NS-2

Simulator for Wireless Sensor Networks ", Canadian Conference on

Electrical and Computer Engineering , Pages: 1372-1375, Provider:

IEEE, 2007.

[60] Chiara Petrioli,Roberto Petroccia, Jon Shusta ,Lee Freitag," From

underwater simulation to at-sea testing using the ns-2 network simulator

", IEEE Journal of OCEANS ,Spain , Pages: 1-9, 2011.

[61] Stéphane Lohier, Abderrezak Rachedi, Erwan Livolant, Ismail

Salhi," Wireless Sensor Network Simulators Relevance compared to a

real IEEE 802.15.4 Testbed ", International Wireless Communications

and Mobile Computing Conference , Pages: 1347-1352, Provider: IEEE,

2011.

[62] Zheng Peng, Jun-Hong Cui, Bing Wang, Keenan Ball, Lee Freitag,"

An Underwater Network Testbed: Design, Implementation and

Measurement", OCEANS MTS/IEEE Conference and Exhibition,

Volume: 2, Pages: 797-802, 2008.

[63] Lee Freitag, Matthew Grund, Sandipa Singh, James Partan, Peter

Koski, Keenan Ball," The WHOI Micro-Modem: An Acoustic

Communications and Navigation System for Multiple Platforms ",

OCEANS, Proceedings of MTS/IEEE , Volume: 1, Pages: 1086-1092,

Sep. 2005.

98

[64] The ns2 simulator manual.

http://www.isi.edu/nsnam/ns/doc/index.html.

[65] The WEP of Aqua-Sim simulator (http://uwsn.engr.uconn.edu).

[66] Tommaso Melodia, Hovannes Kulhandjian, Li-Chung Kuo, Emrecan

Demirors,"Communication Architecture", Book Chapter, Chapter 23,

First Edition, John Wiley & Sons, Inc., 2012.

[67] "Guidelines for Drinking-water Quality", Vol. 1, World Health

Organization, 2004.

[68] Standard methods for the examination of Drinking water in Iraq no.

(417) first update, ICS :13.060.20, 2001.

[69] Nigar A. Aziz, Sabah M. Salih, Nizar Y. Hama-Salh," Pollution of

Tanjero River by Some Heavy Metals Generated from Sewage Wastwater

and Industrial Wastewater in Sulaimani District ", Journal of Kirkuk

University –Scientific Studies, vol.7, No.1, 2012.

http://www.isi.edu/nsnam/ns/doc/index.html

http://uwsn.engr.uconn.edu/

Appendix A

1A-

A. Parameters Definitions and its Source Code

1. Parameters definition

There are five parameters used in performance assessment of protocols

are:-

-Total Energy Consumption: - defined as the total energy consumption

in delivering packet, including transmitting, receiving, and idling energy

consumption of all nodes in the network.

-Average End-to-end Delay: - defined as the average time taken by a

packet for traveling from the source node to the sinks node.

Packet Delivery Ratio: - the ratio of the number of success packets that

received at the sinks to the total number of generated packets at the

source node.

Received Throughput:- In communication networks, such

as Ethernet or packet radio, throughput or network throughput is the

average rate of successful message delivery over a communication

channel. This data may be delivered over a physical or logical link, or

pass through a certain network node. The throughput is usually measured

in bits per second (bit/s or bps), and sometimes in data packets per second

or data packets per time slot.

The system throughput or aggregate throughput is the sum of the data

rates that are delivered to all terminals in a network.

The throughput can be analyzed mathematically by means of queuing

theory, where the load in packets per time unit is denoted arrival rate λ,

and the throughput in packets per time unit is denoted departure rate μ.

http://en.wikipedia.org/wiki/Communication_networks

http://en.wikipedia.org/wiki/Ethernet

http://en.wikipedia.org/wiki/Packet_radio

http://en.wikipedia.org/wiki/Network_node

http://en.wikipedia.org/wiki/Bit

http://en.wikipedia.org/wiki/Data_packets

http://en.wikipedia.org/wiki/Time-division_multiplexing

http://en.wikipedia.org/wiki/Queueing_theory

http://en.wikipedia.org/wiki/Queueing_theory

Appendix A

2A-

Throughput is essentially synonymous to digital bandwidth consumption.

Total Drop Packets: - drop packet or Packet loss occurs when one or

more packets of data travelling across a network fail to reach their

destination. Packet loss is distinguished as one of the three main error

types encountered in digital communications; the other two being bit

error and spurious packets caused due to noise.

2. Source Code of parameters

NS simulator can provide a lot of detailed data on events that occur at the

network. If we wish to analyze the data we may need to extract relevant

information from traces and to manipulate them.

One can of course write programs in any programming language (TCL

and C++) that can handle data files. Yet several tools that seem

particularly adapted for these purposes already exist and are freely

available under various operating systems (unix, linux, windows, etc.).All

they require is to write short scripts that are interpreted and executed

without need for compilation.

Hear, I used awk tool to Processing data or trace file. The awk utility

allows us to do simple operations on data files such as averaging the

values of a given column, summing or multiplying term by term between

several columns, all data- reformatting tasks, etc.

After write the program to compute any parameters or any operation use

the short script to get the result as bellow for example:

awk -f suma.awk Conn4.tr > outfile

http://en.wikipedia.org/wiki/Digital_bandwidth_consumption

http://en.wikipedia.org/wiki/Packet_(information_technology)

http://en.wikipedia.org/wiki/Computer_network

http://en.wikipedia.org/wiki/Bit_error

http://en.wikipedia.org/wiki/Bit_error

Appendix A

3A-

The original file here is Conn4.tr (trace file), the output is written into a

file called oufile and the program language to compute summation is

suma.awk.

The Algorithm of Metrics Calculation is below:-

Step 1: Start.

Step 2: Declare variables recv-size, start-time, stop-time, D-packets,

R-packets and count .

Step 3: Read values event, time, node-id, packet-size, level and seq-no

from trace file.

Step 4: Calculate throughput, E-to-E delay, packet delivery ratio and

dropped packets.

Step 5: Display throughput, E-to-E delay, packet delivery ratio and

dropped packets.

Step 6: Stop.

The source codes of above parameters written in TCL and C++ language are:-

BEGIN {

recvedSize = 0

startTime = 500 (1000,1500,etc)

stopTime = 0

}

{

action = $1

time = $2

node_id = $3

packetsize = $8

level = $4

# Store start time

if (level == "MAC" && event == "s" && pkt_size >= 320) {

Appendix A

4A-

if (time < startTime) {

startTime = time

}

}

# Update total received packets' size and store packets arrival time

if (level == "MAC" && event == "r" && pkt_size >= 320) {

if (time > stopTime) {

stopTime = time

}

# the header

hdr_size = pkt_size % 320

pkt_size -= hdr_size

# Store received packet's size

recvdSize += pkt_size

}

}

END {

printf("Average Throughput[kbps] = %.2f\t\t StartTime = %.2f\t StopTime = %.2f\n",(recvdSize / (stopTime - startTime)) * (8 / 1000),startTime,stopTime)

}

A.1 Throughput Program

BEGIN {

seqno = -1;

droppedPackets = 0;

receivedPackets = 0;

count = 0;

}

Appendix A

5A-

{

if($4 == "MAC" && $1 == "s" && seqno < $6) {

seqno = $6;

}

else if(($4 == "MAC") && ($1 == "r")) {

receivedPackets++;

} else if ($1 == "D" && $7 == "vectorbasedforward" || "vectorbasedvoidavoidance" && $8 > 512){

droppedPackets++;

}

#end-to-end delay

if($4 == "MAC" && $1 == "s") {

start_time[$6] = $2;

} else if(($7 == "vectorbasedforward" || "vectorbasedvoidavoidance") && ($1 == "r")) {

end_time[$6] = $2;

} else if($1 == "D" && $7 == "vectorbasedforward" || "vectorbasedvoidavoidance") {

end_time[$6] = -1;

}

}

END {

for(i=0; i<=seqno; i++) {

if(end_time[i] > 0) {

delay[i] = end_time[i] - start_time[i];

count++;

}

else

Appendix A

6A-

{

delay[i] = -1;

}

}

for(i=0; i<=seqno; i++) {

if(delay[i] > 0) {

n_to_n_delay = n_to_n_delay + delay[i];

}

}

n_to_n_delay = n_to_n_delay/count;

print "\n";

print "GeneratedPackets = " seqno+1;

print "ReceivedPackets = " receivedPackets;

print "Packet Delivery Ratio = " receivedPackets/(seqno+1)*100 "%";

print "Total Dropped Packets = " droppedPackets;

print "Average End-to-End Delay = " n_to_n_delay * 1000 " ms";

print "\n";

}

A.2 Program to compute (packet delivery ratio, drop packet and end

to end delay)

C. Execution Steps of Final Design

The simulation steps are described as blow:

a) First step go to terminal in Linux then to the directory of

final design folder.

Appendix A

7A-

b) Execute the TCL file that contained the final design program and all simulation parameters.

After this we will get the following trace file (some rows as example):-

Appendix A

8A-

c) Then execute the nam command to reveal the network animator as bellow:

Appendix A

9A-

d) To compute any operation or parameters test using the AWK short script to get result that shown in previous section.

Example of throughput result:-

Appendix A

10A-

Also example of other parameters result:-

B. Result Tables

Table (3.2): simulation result of UWSN MAC-protocols

11A-

Broadcast Time 500 1000 1500 2000 2500 3000 3500 4000 4500 Energy(Joule) 74.82062 148.5126 222.5835 295.1154 369.2665 443.178 516.6649 589.8295 664.0498

Throughput(kbps) 3.22 3.23 3.25 3.21 3.21 3.23 3.23 3.22 3.24 Dropped packets 509 950 1332 1984 2469 2750 3239 3712 4016 End to End Delay(ms) 153.753 153.564 153.405 153.72 153.97 153.623 153.559 153.767 153.513

Aloha Time 500 1000 1500 2000 2500 3000 3500 4000 4500 Energy(Joule) 85.91337 170.435447 254.957523 339.479599 424.001675 508.523752 593.045828 677.567904 762.08998

Throughput(kbps) 3.11 3.09 3.09 3.09 3.09 3.08 3.08 3.08 3.08 Dropped packets 0 0 0 0 0 0 0 0 0 End to End Delay(ms)

159.802 159.789 159.785 159.783 159.782 159.781 159.78 159.78 159.78

FAMA Time 500 1000 1500 2000 2500 3000 3500 4000 4500 Energy(Joule) 959.8249 1617.582 2258.656 2950.63 3646.855 4260.158 4979.219 5549.253 6266.268

Throughput(kbps) 2.43 2.4 2.4 2.4 2.44 2.41 2.42 2.39 2.41 Dropped packets 66 103 148 179 241 342 383 432 506 End to End Delay(ms) 87868.8 242720 274359 222260 689742 241090 392483 142719 550535

R-mac Time 500 1000 1500 2000 2500 3000 3500 4000 4500 Energy(Joule) 3.08373 3.687127 4.592366 5.107544 5.503503 6.288133 6.916491 7.949309 8.535994

Throughput(kbps) 0.24 0.24 0.24 0.24 0.24 0.24 0.24 0.24 0.24 Dropped packets 1 301 601 901 1201 1501 1801 2101 2401 End to End Delay(ms) 0 0 0 0 0 0 0 0 0

UW-mac Time 500 1000 1500 2000 2500 3000 3500 4000 4500 Energy(Joule) 146.4337 263.807 374.531 486.5526 586.7821 707.1611 825.8766 933.4548 1034.501

Throughput(kbps) 6.82 6.7 6.45 6.31 6.46 6.45 6.47 6.41 6.33 Dropped packets 57 437 326 517 356 632 823 932 1049 End to End Delay(ms) 233.808 234.169 235.363 236.239 237.51 238.551 237.671 237.552 235.219

B. Result Tables

Table (3.2): simulation result of UWSN MAC-protocols

12A-

5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 737.3131 811.4301 884.4737 959.5056 1031.933 1105.726 1179.471 1252.008 1326.791 1400.042 1473.369

3.21 3.23 3.22 3.23 3.22 3.22 3.23 3.22 3.22 3.22 3.22 4823 5093 5624 5948 6566 7019 7358 8093 8389 8833 9419

153.765 153.711 153.729 153.719 153.693 153.74 153.703 153.68 153.762 153.665 153.753

5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 846.6121 931.1341 1015.656 1100.178 1184.7 1269.222 1353.745 1438.267 1522.789 1607.311 1691.833

3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 3.08 0 0 0 0 0 0 0 0 0

159.779 159.779 159.779 159.779 159.779 159.778 159.778 159.778 159.778 159.778 159.778

5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 6998.068 7722.226 8490.854 9151.564 9470.899 10596.32 11027.4 11540.15 12273.44 13002.13 13644.04

2.43 2.46 2.47 2.46 2.36 2.5 2.43 2.39 2.42 2.44 2.42 576 532 654 864 867 855 986 886 987 845 1157

862170 841886 537052 1085710 1011580 1100200 612293 324033 779453 1017850 1231870

5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 9.141411 9.60985 10.80787 11.28516 12.39845 12.62467 12.91405 13.84731 14.09598 15.45782 15.72977

0.24 0.24 0.24 0.24 0.24 0.24 0.24 0.24 0.24 0.24 0.24 2701 3001 3301 3601 3901 4201 4501 4801 5101 5401 5701

0 0 0 0 0 0 0 0 0 0 0

5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 1132.87 1240.371 1361.888 1489.725 1594.553 1706.756 1789.894 1934.563 2012.031 2111.88 2269.656

6.27 6.19 6.39 6.39 6.37 6.33 6.22 6.3 6.23 6.28 6.41 1134 1332 1157 1554 1614 1957 2031 2461 2211 1979 2259

235.362 235.317 236.258 237.032 237.875 237.488 235.623 236.818 236.338 237.028 236.637

13A-

Table (3.4): Simulation result of UWSN routing protocols

VBF No. of Node 50 100 150 200 250 300 350 400 Energy

Consume (J) 309.388205 600.866545 910.71467

1293.517 1569.489 1929.703 2265.335 2596.261 Packet Delivery Ratio

0.30392 0.58964 0.92832

1 1 1 1 1 Average End to-End Delay(ms)

415.387 422.096 431.336

428.013 413.978 434.387 422.825 435.493 HH-VBF No. of Node 50 100 150 200 250 300 Energy

Consume (J) 581.17853 1180.380196 1741.098426 2285.598934 2991.22854 3469.152206

350 400 Packet Delivery Ratio

0.19392 0.31108 0.48676 0.59976 0.71532 0.87212

4030.934 4604.159 Average End to-End Delay(ms)

675.875 767.922 763.302 676.129 717.509 808.677

1 1 VBVA No. of Node 50 100 150 200 250 300 350 400

Energy Consume (J)

162.25336 271.356 385.754 500.152 614.55 728.948 843.346 957.744

Packet Delivery Ratio

0.20188 0.396 0.596 0.796 0.996 1 1 1

Average End to-End Delay(ms)

860.579 878.22 878.22 878.22 878.22 878.22 878.22 878.22

14A-

Table (3.6): Simulation result of UWSN final design

Final design

Time (sec.) 500 1000 1500 2000 2500 3000 3500 4000 Energy Consume(Joule) 139.915832 243.575887 354.780584 460.687338 549.989824 687.531129 776.798682 880.160729 Throughput (kbps) 6 5.94 6.39 6.34 6.25 6.46 6.31 6.31 Dropped Packets 147 280 238 359 264 535 588 650

End to End Delay 237.682 237.27 234.706 236.794 237.653 236.72 237.265 236.442

Cont.

4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 980.3803

69 1108.60

97 1172.524

66 1296.773

68 1416.313

96 1488.375

17 1621.901

69 1730.895

63 1833.279

29 1916.39

1 2048.317

89 2165.070

71 6.24 6.33 6.05 6.28 6.32 5.99 6.09 6.2 6.34 6.14 6.32 6.31 899 1125 1273 1142 1253 1954 2228 1981 1517 1912 1848 1988

237.398 236.635 236.65 237.724 236.388 235.397 236.538 236.274 237.235 236.046 236.485 236.048

Appendix B

15B-

Publication arises from project

1. Mohammed N. and Ahmed M. "Performance Assessment of MAC

Layer Protocols Based on Underwater Wireless Sensor Networks ",

Journal of Global Research in Computer Science, Volume 4, No. 3,

March 2013.

2. Mohammed N. and Ahmed M. " Comparative Study of Routing

Protocols in Pollution Monitoring System Based on Underwater

Wireless Sensor Networks " Oriental Journal of Computer Since &

Technology, Vol. 6, No. 2, June 2013

الخالصة

إلى ان تأخذ حيزا في مجالنشطة التي تحتاج يصف نظام رصد التلوث العمليات و األ

نوعية بيئة معينة .مراقبة و وصف

مجالين مهمين خاللو قدمت من تم تمثيلها العمليات الفعاليات و ،فإن في هذه األطروحة و

في جديد للمياه مؤشر جودة تطويروفي المياه مغمورةال، تصميم شبكة االستشعار الالسلكية هما

.FISراق باستخدام نظام الالع

من بروتوكوالتخمسة فياألكبر منها على تقييم األداء الجزء فيتركز هذه األطروحة

، و Broadcast MAC ،ALOHA ،R -MAC ،FAMA(هي UWSN MACال طبقة

UWAN_MAC ( اإلنتاجية ثم ترشيح استالمحيث استهالك الطاقة وئية من البيئة المافي

التأخير بين و الضائعة الحزم مجموعم السابق و مقارنتها من حيث ثالثة منهم بناء على التقيي

بروتوكول دى متانةللحقائق حول م فحص واستقصاء. عالوة على ذلك، قدمت طرف وآخر

من خالل مقارنة بين المغمورة في المياه ت االستشعار الالسلكيةفي شبكا تهطاق كفاءةوالتوجيه

) . وكان VBVA( ) و VBF (، )HHVBF( وهيالتوجيه تبروتوكوالمن اهم ثالثة

و بين النهايات التأخير تقييم أداء هذه البروتوكوالت باستخدام مقاييس استهالك الطاقة ، متوسط

في Aqua-Sim برنامج محاكاة يسمى . يتم التقييم عن طريق استخدام نسبة تسليم الحزم

في وتم العمل عليها NS2المحاكاة مثبت على برنامج المغمورة في المياهشبكات االستشعار

بيئة لينكس .

ثات لمراقبةإلكمال نظام ا تم عمل مؤشر متين ومرن ليستخدم كأساس لكل الملو

الى تطوير مؤشر فيهدلتقديم عمل FISالوتم عمل ذلك بإستخدام نظام ، المراد مراقبتها

التقليدية بيئيةالمؤشرات ال بحيث يستخدم المنطق الضبابي لتطويرالعراق ، المياه في جودة

وكذلك لتخطي مياه الشرب نوعية المياه ، وخاصة بالنسبة لل للقضاء على الروتين في تقييم

قترح مع الظروف الم. لتقييم أداء المؤشر التي تتمثل بعدم الدقة والموضوعية المشاكل الروتينية

يانات نوعية ، العراق ، وقد استخدمت ب نهر دجلةسة الحالة التي أجريت على درا ،فانالحقيقية

الطريقة المقترحة من تحققت التي النتائج أن تبينمختلفة . قع ااخذت من مو ت المياه من عينا

العادية طريقةال من عليها الحصول تم التي النتائج من أقل هي) FWQI( المياه جودة لمؤشر

.محددة لفترةو ٪ 41.64 بحوالي التلوثمقياس من النهائية الدرجة في

وزارة التعليم العالي والبحث العلمي

الجامعة التكنولوجة

قسم هندسة الحاسوب

رســـالة

كجزء قسم هندسة الحاسوب في الجامعة التكنولوجيةمقدمة الى

الحاسوبفي هندسة علومماجستيرال شهادةمن متطلبات نيل

من قبل

احمد موسى ديناراحمد موسى دينار

)2007(بكالوريوس

بإشراف

األستاذ المساعد الدكتور محمد نجم عبدهللا

1435محرم تشرين الثاني 2013

على على في االنهار العراقيةفي االنهار العراقية التلوثالتلوث مراقبةمراقبةنظام نظام المغمورة في المغمورة في تشعار الالسلكية تشعار الالسلكية أساس شبكات االسأساس شبكات االس

الماءالماء

Documents

Iraqi Rivers Pollution Monitoring System Based on Underwater Wireless Sensor Networks