Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
i
EAST: Energy-efficient Adaptive Scheme for Transmission in Wireless Sensor Networks
By Mr. Muhammad Tahir
Registration Number: CIIT/FA10-REE-042/ISB MS Thesis
In Electrical Engineering
COMSATS Institute of Information Technology Islamabad – Pakistan
FALL, 2012
ii
EAST: Energy-efficient Adaptive Scheme for
Transmission in Wireless Sensor Networks
A Thesis presented to COMSATS Institute of Information Technology
In partial fulfillment of the requirement for the degree of
MS (Electrical Engineering)
By
Mr. Muhammad Tahir
CIIT/FA10-REE-042/ISB
Fall, 2012
iii
EAST: Energy-efficient Adaptive Scheme for Transmission in Wireless Sensor Networks
A Graduate Thesis submitted to Department of Electrical Engineering as partial fulfillment of the requirement for the award of Degree of M. S.
(Electrical Engineering).
Name Registration Number Mr. Muhammad Tahir CIIT/FA10-REE-042/ISB
Supervisor: Dr. Nadeem Javaid, Assistant Professor,
Center for Advanced Studies in Telecommunications (CAST), COMSATS Institute of Information Technology (CIIT),
Islamabad Campus, December, 2012
iv
Final Approval
This thesis titled
EAST: Energy-efficient Adaptive Scheme for Transmission in Wireless Sensor Networks
By Mr. Muhammad Tahir
CIIT/FA10-REE-042/ISB
has been approved for the COMSATS Institute of Information Technology, Islamabad
External Examiner: __________________________________ (To be decided)
Supervisor: ________________________ Dr. Nadeem Javaid /Assistant professor, Center for Advanced Studies in Telecommunications (CAST), CIIT, Islamabad.
Head of Department:________________________ Dr. Raja Ali Riaz / Associate professor, Department of Electrical Engineering, CIIT, Islamabad.
v
Declaration
I Mr. Muhammad Tahir, CIIT/FA10-REE-042/ISB herebyxdeclare that I havexproduced the workxpresented inxthis thesis, duringxthe scheduledxperiod of study. I also declare that I havexnot taken anyxmaterial from anyxsource exceptxreferred toxwherever due that amountxof plagiarism isxwithin acceptablexrange. If a violationxof HEC rulesxon research hasxoccurred in thisxthesis, I shall be liablexto punishablexaction under the plagiarismxrules of the HEC.
Date: ________________ ________________ Mr. Muhammad Tahir CIIT/FA10-REE-042/ISB
vi
Certificate
It is certified that Mr. Muhammad Tahir, CIIT/FA10-REE-042/ISB has carried out all the work related to this thesis under my supervision at the Department of Electrical Engineering, COMSATS Institute of Information Technology, Islamabad and the work fulfills the requirements for the award of MS degree.
Date: _________________
Supervisor:____________________ Dr. Nadeem Javaid /Assistant professor, Center for Advanced Studies in Telecommunications (CAST), CIIT, Islamabad.
____________________________ Head of Department: Dr. Raja Ali Riaz/Associate Professor, Department of Electrical Engineering, CIIT, Islamabad.
vii
DEDICATION
Dedicated to my parents.
viii
ACKNOWLEDGMENT I am heartily grateful to my supervisor, Dr. Nadeem Javaid, whose patient encouragement, guidance and insightful criticism from the beginning to the final level enabled me have a deep understanding of the thesis. Lastly, I offer my profound regard and blessing to everyone who supported me in any respect during the completion of my thesis especially my friends in every way offered much assistance before, during and at completion stage of this thesis work.
Mr. Muhammad Tahir CIIT/FA10-REE-042/ISB
ix
Abbreviations and Notations
WSN Wireless s Sensor Network 𝑅𝑆𝑆𝐼𝑙𝑜𝑠𝑠 Transmitter Power Loss ECC Error Control Code 𝐸𝑏/𝑁0 Bit Energy per Noise Power
Spectral Density ACK Acknowledgment LMA Local Mean Algorithm LifeMsg Life Message LifeAckMsg Life Acknowledgment Message NodeMinThresh Nodes Minimum Threshold NodeMaxThresh Nodes Maximum Threshold LINT Local Information No Topology LILT Local Information Link-state
Topology DTPC Dynamic Transmission Power
Control PRR Packet Reception Ratio ATPC Adaptive Transmission Power
Control ARQ Automatic Repeat Request FEC Forward Error Correction CRC Cyclic Redundancy Check HARQ Hybrid ARQ 𝑛𝑑 Desired Number of Neighbor
Nodes 𝑛𝑐 Current Number of Neighbor
Nodes E(t) Error 𝑃𝑙𝑒𝑣𝑒𝑙 Transmitter Power Level T Temperature D Distance Between Each Node SNR Signal Power to Noise Power
x
Abbreviations and Notations
ɳ Spectral Efficiency F Frequency RNF Receiver Noise Figure R Data Rate B Bandwidth Λ Wavelength K Boltzman Constant M Noise Proportionality Constant 𝑃𝑡 Transmitter Power Σ Standard Deviation of Signal PL Path Loss P Pressure H Humidity 𝐸𝐶𝐶𝑔𝑎𝑖𝑛 Coding Gain 𝑃𝑟 Receiver Power 𝑃𝑒 Probability of Error O-QPSK Offset - Quadrate Phase Shift
Keying DSSS Direct Sequence Spread Spectrum 𝑃𝑛𝐻 Transmitter Power for n Hop RS Reed Solomon CC-Hard decision
Convoloutional Code Hard decision
CC-Soft decision
Convoloutional Code Soft decision
BER Bit Error Rate IEEE802.15.4 WSN Standard
xi
List of Publications
[1] Manzoor. B, Javaid. N, Bibi. A, Tahir. M, Khan. Z. A, “Noise Filtering, Channel Modeling and Energy Utilization in Wireless Body Area Networks”, 3rd International Symposium on Advances in Embedded Systems and Applications (ESA2012) in conjunction with 9th IEEE International Conference on Embedded Software and Systems (ICESS-2012), Liverpool, UK, 2012.
[2] M.Tahir, N. Javaid, “EAST:Energy-efficient Adaptive Scheme for Transmission in Wireless Sensor Networks”, submitted in 10th IEEE International Conference on Wireless On-demand Network Systems and Services (WONS'13), March 18-20, 2013, Banff, Canada.
[3] M.Tahir, N. Javaid, “EETS:Energy Efficient Transmission Scheme for Wireless Sensor Networks”, submitted in 4th IEEE International Conference on Ambient Systems, Networks and Technologies (ANT-13) June 25-28, 2013, Halifax, Nova Scotia, Canada.
xii
ABSTRACT
One of the major challenges in design of Wireless Sensor Networks (WSNs) is to reduce energy consumption of sensor nodes to prolong lifetime of finite-capacity batteries. In this thesis, we propose Energy Efficient Adaptive Scheme for Transmission (EAST) in WSNs. EAST is IEEE 802.15.4 standard compliant. In this approach, open-loop is used for temperature-aware link quality estimation and compensation. Whereas, closed-loop feedback help to divide network into three logical regions to minimize overhead of control packets on basis of Threshold transmitter power loss (𝑅𝑆𝑆𝐼𝑙𝑜𝑠𝑠) for each region and current number of neighbor nodes that help to adapt transmit power according to link quality changes due to temperature variation. Simulation results show that propose scheme; EAST effectively adapts transmission power to changing link quality with less control packets overhead and energy consumption compared to classical approach with single region in which maximum transmitter power assigned to compensate temperature variation. We have also shown that how in presence of existing Error Control Coding (ECC) techniques and decoder complexity energy efficiency increased. That is by estimating transmitter power for each sensor node in given environment. Since adoption of ECC reduces required transmitter power for reliable communication, while increase processing energy of decoding operations. Required transmitter power for sensor nodes in given environment for different coding techniques like reed-solomon (RS), convolutional (CC) energy efficiency and bit error rate has been analyzed for different Eb/N0.
Contents
1 Introduction 1
1.1 Wireless Sensor Networks . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Error Control Coding . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Multi-Hop Transmission . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Related Work and Motivation 4
2.1 Transmission Power Control Techniques . . . . . . . . . . . . . . . . 4
2.2 Error Correction Techniques . . . . . . . . . . . . . . . . . . . . . . 5
2.3 Multi-Hop Transmission Techniques . . . . . . . . . . . . . . . . . . 6
3 Energy Efficient Transmission in Wireless Sensor Networks 7
3.1 EETS:Energy Efficient Transmission Scheme . . . . . . . . . . . . . 7
3.2 EAST:Energy-efficient Adaptive Scheme for Transmission . . . . . . 10
3.3 MEAST:Multi-Hop Energy-efficient Adaptive Scheme for Transmis-
sion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4 Results and Discussion 18
4.1 Simulation Results of EETS . . . . . . . . . . . . . . . . . . . . . . 18
4.2 Simulation Results of EAST . . . . . . . . . . . . . . . . . . . . . . 20
4.3 Simulation Results of MEAST . . . . . . . . . . . . . . . . . . . . . 25
5 Conclusion 33
References 40
xiii
List of Figures
3.1 Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2 Flow Chart of Reference Node . . . . . . . . . . . . . . . . . . . . . 16
3.3 Transmission distances for: (a) signle-hop, (b) double-hop, (c) triple-
hop, (d) quad-hop. . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.1 Required Transmitter power versus distance . . . . . . . . . . . . . 18
4.2 Transmitter power versus distance for estimated fading . . . . . . . 19
4.3 Transmitter power versus distance with and without fading . . . . . 20
4.4 Transmitter power versus distance for various environments . . . . . 21
4.5 Transmitter power versus distance for various channel coding tech-
niques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.6 Transmitter energy versus distance for different coding techniques . 23
4.7 Bit error rate versus given Eb/No . . . . . . . . . . . . . . . . . . . 24
4.8 Temperature for different Nodes . . . . . . . . . . . . . . . . . . . . 25
4.9 RSSI-loss for different Nodes . . . . . . . . . . . . . . . . . . . . . . 26
4.10 Power level for different Nodes . . . . . . . . . . . . . . . . . . . . . 26
4.11 Transmitter Power for different Nodes . . . . . . . . . . . . . . . . . 27
4.12 Power level for different regions . . . . . . . . . . . . . . . . . . . . 27
4.13 Power level using EAST for different regions . . . . . . . . . . . . . 28
4.14 Power level save for region A . . . . . . . . . . . . . . . . . . . . . . 28
4.15 Power level save for region B . . . . . . . . . . . . . . . . . . . . . . 29
4.16 Power level save for region C . . . . . . . . . . . . . . . . . . . . . . 29
4.17 Power level save in region A for different Reference Node Locations 30
4.18 Power level save in region B for different Reference Node Locations 30
4.19 Power level save in region C for different Reference Node Locations 31
4.20 Power level for region A in different environments . . . . . . . . . . 31
4.21 Power level for region B in different environments . . . . . . . . . . 32
4.22 Power level for region C in different environments . . . . . . . . . . 32
xiv
List of Tables
1.1 IEEE802.15.4 Specifications . . . . . . . . . . . . . . . . . . . . . . 1
3.1 Input Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.2 Typical values of path exponent . . . . . . . . . . . . . . . . . . . . 14
4.1 Simulation Parameters . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2 Estimated Parameters . . . . . . . . . . . . . . . . . . . . . . . . . 22
xv
Chapter 1
Introduction
WSNs are being considered for many applications, including industrial, security
surveillance, medical, environment and weather monitoring. Due to limited bat-
tery lifetime at each sensor node minimizing transmitter power to increase energy
efficiency and network lifetime is useful for reliable network operation. Limited
battery lifetime requires low power sensing, processing and communication sys-
tem. In WSN sensor nodes must consume minimum amount of transmitter power
that is needed to provide energy efficient communication. Sensor nodes consist of
three parts sensing unit, processing unit and transceiver [1].
1.1 Wireless Sensor Networks
In WSNs, sensor nodes are widely deployed in different environments to collect
data. Because sensor nodes usually operate on limited battery energy efficiency is
an important factor. Each sensor node communicate using a low power wireless
link and its link quality varies significantly due to environmental dynamics like
temperature etc. Therefore, while maintaining good link quality with its neighbors
we need to reduce energy consumption for data transmission to extend network
lifetime [2]. Specifications of WSN standard IEEE802.15.4 given as, (a) Frequency
2.45 GHz (b) Channels Up to 16 (c) Data rate 250 kbps (d) Bandwidth 83.5 MHz.
Table 1.1: IEEE802.15.4 Specifications
IEEE802.15.4
Frequency Channels Data rate Bandwidth2.45GHz Up to 16 250kbps 83.5MHz
1
1.2 Error Control Coding
ECC use in different applications, such as communications and memory storage
systems. That increases their corresponding data reliability by adding properly
estimated redundancy to main stream data. In wireless communication, given
BER can be achieved at lower transmitter power with use of ECC at cost of band-
width as well as additional encoding-decoding complexity at transceiver. That
complexity is critical where concern is towards bandwidth efficiency. In WSN,
if associated processing power is greater than coding gain then applying ECC is
not energy-inefficient. Since distance between neighbor nodes is in range of few
meters, computational power can be comparable to transmission power [3].
ECC is actually classic approach used to increase link reliability and lower required
transmitter power. However, lower power at transmitter comes at cost of extra
power consumption due to decoder complexity at receiver. Stronger codes provide
better performance with higher power consumption than simple error control codes
[4]. New approach for transmitter design of sensor nodes has been propose to lower
required transmitter power. In this approach, estimation of multi-path fading
effect on signal for given environment help to lower transmitter power, otherwise
if we take average value of fading effect then most of times we have to transmit
extra power that minimizes WSN efficiency [5].
1.3 Multi-Hop Transmission
Network coverage area is often much larger then radio range of single node, so
in order to reach some destination node can use other nodes as relays. This
type of communication is known as multi-hop communication.Overall WSN node
power consumption depends on processors, transceivers power consumption and on
the operation regime of these components (switching between idle and operating
mode). Most of the node energy is consumed by radio transmission. Power savings
in radio transmission are usually achieved by use of energy efficient medium access
and routing protocols. The most modern radio transceivers could adjust their
transmitting power, so some destination could be reach with either large number
of smaller hops (multi-hop) or small number of larger hops (single-hop) [6].
Energy efficiency of these two approaches depends on path loss between trans-
mitter and receiver and power consumption of the radio transceiver in various
operating modes. It is theoretical known from state of the art that multi-hop
communication is more efficient then single-hop communication. This is opposite
2
to observations in some real world WSN, which shows that single-hop communi-
cation, can be much more energy efficient then multi-hop communication. Besides
energy efficiency, single-hop communication can also have advantages for other
network parameters, such as end-to-end delay, lower packet loss, etc [7].
To efficiently compensate link quality changes due to temperature variations, I
propose a new scheme for transmission power control EAST that improves energy
efficiency while achieving required reliability between sensor nodes. New scheme
base on combination of open-loop and closed-loop feedback processes in which
I divide network into three regions on basis of Threshold RSSIloss. In open-
loop process, each node estimates link quality using its temperature sensor [8].
Estimated link quality degradation is then effectively compensated using closed-
loop feedback process by applying propose transmission power control scheme. In
closed-loop feedback process, appropriate transmission power control is obtained
which assign substantially less power than those required in existing transmission
power control schemes.
As we know that required transmitter power for any sensor node in WSN depends
upon distance between sensor nodes, frequency that is used for transmission, data
rate and required signal to noise ratio at receiver if we consider free space path
loss model [9]. Due to multi-path fading effect on signal in free space we need to
transmit more power to detect and receive signal at receiver reliably that consumes
more power and limit battery life time. Normally we take value of multi-path fad-
ing effect on average (4dB-10dB) in our approach I estimate that effect for given
environment, frequency and elevation angle, so that we accurately estimate trans-
mitter power at each sensor node. International Telecommunication Union model
help to estimate multi-path fading effect for given environment [10]. Required
transmitter power for sensor nodes in given environment for different coding tech-
niques like Reed Solomon, Convolution codes energy efficiency and bit error rate
has been analyzed for different Eb/N0.
3
Chapter 2
Related Work and Motivation
To transmit data efficiently over wireless channels in WSNs, existing schemes set
some minimum transmission power for maintaining reliability. These schemes ei-
ther decrease interference among nodes or unnecessary energy consumption. In
order to adjust transmission power, reference node periodically broadcasts a bea-
con message. When neighbor nodes hear a beacon message from a reference node,
neighbor nodes transmit an ACK message. Through this interaction, reference
node estimate connectivity between neighbor nodes [1].
2.1 Transmission Power Control Techniques
In Local Mean Algorithm (LMA), a reference node broadcasts LifeMsg message.
Neighbor nodes transmit LifeAckMsg after they receive LifeMsg. Reference nodes
count number of LifeAckMsgs and transmission power is controlled by maintaining
appropriate connectivity. For example if number of LifeAckMsgs is less than
NodeMinThresh transmission power is increased [11]. In contrast, if number of
LifeAckMsgs is more than NodeMaxThresh transmission power is decreased. As
a result, they provide improvement of network lifetime in a sufficiently connected
network. However, LMA only guarantees connectivity between nodes and cannot
estimate link quality [12].
Local Information No Topology/Local Information Link-state Topology (LINT/LILT)
and Dynamic Transmission Power Control (DTPC) uses RSSIloss to estimate
transmitter power level. Nodes exceeding Threshold RSSIloss are regarded as
neighbor nodes with reliable links. Transmission power also controlled by Packet
Reception Ratio (PRR) metric [13].
Since RSSIloss is directly proportional to temperature. Adaptive Transmission
4
Power Control (ATPC) adjusts transmission power dynamically according to spa-
tial and temporal effects. This scheme tries to adapt link quality that changes over
time by using closed-loop feedback. However, in large-scale WSNs it is difficult to
support scalability due to serious overhead required to adjust transmission power
of each link [11]. Existing approaches estimate variety of link quality indicators
by periodically broadcasting a beacon message. In addition, feedback process is
repeated for adaptively controlling transmission power. In adapting link quality
for environments where temperature variation occur, packet overhead for trans-
mission power control should be minimized. Reducing number of control packets
while maintaining reliability is an important technical issue [14].
New transmission power control scheme for energy efficient transmission EAST
help to efficiently compensate link quality changes due to temperature variation
[15]. To reduce packet overhead for adaptive power control temperature measured
by sensors is utilized to adjust transmission power level for all three regions based
on RSSIloss compared to single region in which large control packets overhead
occur even due to small change in link quality. Closed-loop feedback process is
additionally executed to minimize control packets overhead and required trans-
mitter power level [16].
2.2 Error Correction Techniques
To enhance link reliability for sending data on channel, techniques such as ARQ
(Automatic Repeat Request) and FEC (Forward Error Correction) are employed
[16]. FEC introduce error correcting techniques to counter bit errors by adding
redundancy (extra bits) in information packets. Receiver use these extra bits to
detect and correct errors due to channel imparients. FEC coding techniques are
related with unnecessary overhead that decreases energy efficiency when channel
is relatively error free. In ARQ technique for retransmission only error detec-
tion capability is given; receiver requests to transmitter retransmission of packets
received in error.
Normally ARQ scheme uses Cyclic Redundancy Check (CRC) codes to detect
errors. At receiver CRC verifies packet. If it detects errors, node asks for retrans-
mission to transmitter that is negative acknowledgement. If reception is correct,
a positive acknowledgement is sent to transmitter node. Hybrid ARQ schemes
be developed using combination of FEC and ARQ. Typical error control coding
techniques for WSNs are discussed in [17].
New approach for transmitter design of sensor nodes has been proposed to lower
5
required transmitter power. In this approach, estimation of multi-path fading
effect on signal for given environment help to lower transmitter power, otherwise if
we take average value of fading effect then most of times we have to transmit extra
power that minimizes WSN efficiency [16]. ECC is actually classic approach used
to increase link reliability and lower required transmitter power. However, lower
power at transmitter comes at cost of extra power consumption due to decoder
complexity at receiver. Stronger codes provide better performance with higher
power consumption than simple error control codes [18]. Required transmitter
power for sensor nodes in given environment for different coding techniques like
Reed Solomon, Convolution codes energy efficiency and bit error rate has been
analyzed for different Eb/N0.
2.3 Multi-Hop Transmission Techniques
A strong error control coding technique correct many errors and as a result energy
consumed too high for an energy constrained sensor network. Same error control
technique for whole network could be a good choice in some cases, not always.
Most of the times it is necessary to apply best available error control coding
technique. Error control technique selection based on number of hops packet
traveled within network [19]. If sensor node has to send data packet to sink node,
before packet reaches its destination it travel through some other nodes of sensor
network [18]. If packet gets lost at first hop, only energy to send packet from a
sensor to a specific node is lost. If packet is corrupted after few more hops, much
more energy be spent to transmit packet through network. In this sense, a packet
is more important if it travels through more nodes in network, and consequently,
more energy is being consumed [2].
Issue of sending packets over long hops or short hops has been raised by many
authors in recent past and their conclusions are varied depending on approach
taken considered. Methods of transmission energy minimization consists of de-
creasing the transmission range of each node. This scheme will reduce overall
power consumption of sensor network, a route with many short hops is generally
more energy-efficient than one with a few long hops. There are many reasons why
long-hop communication is more advantageous. One of them is the power effi-
ciency. Authors claimed that although the transmitted energy drops significantly
with distance, the reduction of radiated power does not yield a decrease in the
total energy consumption.
6
Chapter 3
Energy Efficient Transmission in
Wireless Sensor Networks
3.1 EETS:Energy Efficient Transmission Scheme
Estimation of transmitter power for sensor nodes on basis of given meteorological
conditions and how in presence of existing error control coding techniques and
decoder complexity energy efficiency be increased. To increase battery life time
of sensor nodes we need exact estimated transmitter power so that we consume
power efficiently and communicate reliably. For that we need to know distance
between transmitter and receiver sensor nodes, frequency used for communica-
tion and multi-path fading effect. Estimation of multi-path fading effect due to
meteorological conditions for given environment given as [17]:
χσ = a(p)× σdB (3.1)
Here σ is standard deviation of signal for considered period and propagation path
that depends upon on reference standard deviation of signal, antenna averaging
factor, frequency, elevation angle of antenna and a(p) that is time percentage
factor. χσ gives cumulative distribution of fading with respect to time. For esti-
mation of reference standard deviation of signal we need wet-term of refractivity
that depends upon given meteorological conditions (T, P, H). For free space model
path loss be estimated as given by Friis law [20]:
PL(d) = PL(d0) + n.10.log10(d/d0) + χσdB (3.2)
7
Here χσ (multi-path effect) that actually attenuate signal and decrease signal level,
d0 reference distance, d is distance between two sensor nodes, n is for different
environments and PL(d0) is reference path loss. Path loss also be defined as ratio
of received signal power to transmitted power as given below [21]:
PL(d) = SRX(d)/PTX (3.3)
Received signal power to noise power ratio depends upon data rate R bandwidth
B and given Eb/N0.
S/N = (R× Eb)/(N0 ×B) (3.4)
From that we estimate minimum required transmitter power for given distance,
frequency, received signal strength in terms of Eb/N0 and bandwidth [22]:
PTX = ηEb/N0mKTB(4πd/λ)2 (3.5)
Here η is spectral efficiency, m noise proportionality constant, k is Boltz-man
constant. Noise proportionality constant actually antilog of Receiver Noise Fig-
ure(RNF). Receiver power defined as difference between transmitted power and
path loss [23]:
PRX = PTX − PL(d)dB (3.6)
For coded system transmitted power be estimated as given below, where ECCgain
is coding gain due to different coding techniques that actually minimizes required
transmitter power. Transmission energy is just required transmitter power divide
by data rate [22]:
PTX,ECC = (PTX)/(10ECCgain/10) (3.7)
EbTX= PTX/R (3.8)
ECCgain be computed as difference between un-coded SNR and coded SNR.
When higher order coding used then coding gain increases and required power
decreases.
8
Table 3.1: Input Parameters
Frequency(f) 400MHz-10GHzDistance(d) 1-100mVoltage(v) 1.2-3.7v
Environment(n) 2,3,4Temperature(T) 300K
RNF(Receiver Noise Figure) 5dBEb/N0 8.3dB
ECCgain(Coding gain) 1,2.3,4.2dBAntenna(Planar) 16*5mm
Algorithm 1 Proposed Algorithm
T ← TemperatureP ← PressureH ← Humidityθ← Anglef ← Frequencyg(x)← Antenna averaging factorσ ← Standard deviation of signalχσ ← Fadingd← Distance between each nodePL← Path lossPr ← Received powerPt ← Transmitter power
ECCgain = SNRU − SNRECC (3.9)
For convolutional code bit error rate is given that based on hard or soft decision
[24]:
Pe = Q(√
(5Eb/N0)/(1− 2exp(−Eb/2N0)2)) (3.10)
For reed solomon codes [25]:
Pb = 2(m−1)/((2m)− 1)× Pe (3.11)
9
3.2 EAST:Energy-efficient Adaptive Scheme for
Transmission
In this section, we present energy efficient adaptive transmission power control
scheme for energy efficient transmission that maintain link quality during temper-
ature variation in wireless environment. Our transmission power control scheme
is designed to efficiently combine closed-loop and open-loop feedback processes to
divide network into three logical regions. It utilizes open-loop process based on
sensed temperature information to reduce overhead for transmission power control
according to temperature variation. Closed-loop feedback process based on con-
trol packets is further used to accurately adjust transmission power. By adopting
both open-loop and closed-loop feedback processes we divide network into three
regions, A for High RSSIloss, B for Medium RSSIloss and C for Low RSSIloss on
basis of Threshold RSSIloss for each region.
Power
ControllerEAST Network
Temperature
Open Loop
Closed Loop
+/-Nd(t) Nc(t)
Figure 3.1: Block Diagram
In order to assign minimum and reachable transmission power to each link EAST
is designed. EAST has two phases that is initial and run-time. In initial phase
each node build a model for each of its neighbors links. In run-time phase based
on previous model EAST adapt the link quality to dynamically maintain each link
with respect to time. In a relatively stable network, control overhead occurs only
in measuring link quality in initial phase. In a relatively unstable network because
link quality is continuously changing initial phase is repeated and serious overhead
occur. Before we present block diagram for our propose scheme some variables are
defined as follows (1)Number of current neighbor nodes nc(t) (2) Desired number
of neighbor nodes nd(t) (3)Error: e(t) = nd(t)−nc(t),(4)Transmission power level
Plevel.
Fig3.1 shows system block diagram of our propose scheme. In order to adjust
transmission power, transmission power level determined as connectivity with
neighbor nodes. After comparing number of current neighbor nodes with a set
10
point desired number of neighbor nodes power controller adjusts transmission
power level accordingly. PRR, ACK, and RSSIloss used to determine connectiv-
ity. ACK estimate connectivity, it cannot determine link quality. PRR estimate
connectivity accurately that causes significant overhead [26]. In our scheme, we
use RSSIloss for connectivity estimation, which measure connectivity with rela-
tively low overhead.
Power controller adjust transmission power level by utilizing both number of cur-
rent neighbor nodes and temperature sensed at each neighbor node. Since power
controller is operated not merely by comparing number of current neighbor nodes
with desired number also by using temperature-compensated power level, so that
it reach to desired power level rapidly. If temperature is changing then tempera-
ture compensation is executed on basis of relationship between temperature and
RSSIloss. Network connectivity maintained with low overhead by reducing feed-
back process between nodes which is achieved due to logical division of network
while link quality is changing due to temperature variation [27].
Transmission power loss due to temperature variation formulated using relation-
ship between RSSIloss and temperature experimented in Bannister et al.. Math-
ematical expression for RSSIloss due to temperature variation is as follows [9]:
RSSIloss[dBm] = 0.1996 ∗ (T [Co]− 25[Co]) (3.12)
To compensate RSSIloss estimated from Eq.(3.12) we have to control output power
of radio transmitter accordingly. Relationship between required transmitter power
level and RSSIloss is formulated by Eq.(3.13) using least square approximation
[9]:
Plevel[dBm] = [(RSSIloss + 40)/12]2.91 (3.13)
Based on Eqs (3.12, 3.13), we obtain appropriate power level to compensate
RSSIloss due to temperature variation. To compensate path loss due to dis-
tance between each sensor node in WSN free space model help to estimate actual
required transmitter power. After addition of required power level due to tem-
perature variation and distance given in Eq.(3.14), we estimate actual required
transmitter power between each sensor node. For free space path loss model we
need number of nodes, distance between each node, required Eb/No depends upon
SNR, spectral efficiency η, frequency f and receiver noise figure (RNF ) required
[28]:
11
Pt[dBm] = [η ∗ (Eb/N0) ∗mkTB ∗ (4πd/λ)2 +RNF ] + Plevel (3.14)
Our scheme aims to simplify transmission power control by compensating RSSIloss
change based on temperature information sensed at each node. Propose compen-
sation scheme does not require any communication overhead with neighbor nodes,
rather utilizes information gathered from temperature sensor. Open-loop control
reduce significantly complexity of closed-loop feedback control for transmission
power control. We define important parameters for our propose scheme,(1)RSSIloss
Threshold for each region. (2)Number of desired neighbor nodes in each region
nd(t) = nc(t)− 5, (3)Transmission power level for each region [29].
Algorithm 2 EAST Algorithm
1: r ← Number of rounds2: N ← Number of nodes in Network3: d← Distance between each neighbour node and reference node4: T ← Temperature for each node5: RSSIloss ← Transmission power loss for each node6: Plevel ← Power level for each node7: Pt ← Transmitter power for each node8: HighRSSIloss ← Region A9: MediumRSSIloss ← Region B10: LowRSSIloss ← Region C11: Ncurrent ← Current number of nodes12: Ndesired ← Desired number of nodes13: if RSSIloss(A,B,C) ≥ RSSIloss(Threshold) then14: if NCurrent(A,B,C) ≥ NDesired(A,B,C) then15: RSSIloss(new)(A,B,C) = RSSIloss(Threshold)16: else17: RSSIloss(new)(A,B,C) = RSSIloss(A,B,C)18: end if19: end if20: if RSSIloss(A,B,C) < RSSIloss(Threshold) then21: RSSIloss(new)(A,B,C) = RSSIloss(A,B,C)22: end if23: Plevsl(Save)(A,B,C) = Plevel − Plevel(new)(A,B,C)
Threshold RSSIloss is minimum value required to maintain link reliability. Ref-
erence node broadcasts beacon message periodically to neighbor nodes and wait
ACKs. If ACKs are received from neighbor nodes then RSSIloss is estimated for
logical division of network, number of nodes with high RSSIloss considered in re-
gion A, medium RSSIloss considered in region B, and with low RSSIloss in region
C. If (RSSIloss ≥ RSSIloss Threshold) and (Ncurrent ≥ Ndesired) then Threshold
transmitter power level assigned if for similar case (Ncurrent < Ndesired) then sim-
12
ilar transmitter power assigned and if (RSSIloss < RSSIloss Threshold) then by
default keep same transmitter power level. Given below is complete algorithm for
EAST.
Fig3.2 shows complete flow chart of reference node. Neighbor nodes receive beacon
message from reference node. Then neighbor node senses temperature by using
locally installed sensor and checks if temperature change. If there is any tem-
perature change, compensation process is executed on basis of Eqs (3.12, 3.13).
Neighbour node send an ACK message including temperature change information
with a newly calculated power level. Applying this temperature-aware compen-
sation scheme we reduce overhead caused by conventional scheme in changing
temperature environments .
3.3 MEAST:Multi-Hop Energy-efficient Adaptive
Scheme for Transmission
Radio channel between transmitter and receiver can be established only when
strength of the received radio signal is grater then receivers sensitivity threshold.
The reduction in signal power density, on the path between transmitter and re-
ceiver, is called path loss. Realistic path loss modeling can be a very complex
task because transmitted radio waves could be reflected, absorbed or scattered by
the obstacles. Receivers in a real environment receive not one but many delayed
components of the original signal. Such phenomenon is called multi-path fading.
The simplest path-loss model, called free-space, assumes that there are no ob-
structions between transmitter and receiver. Free-space path loss is proportional
to the square of the distance between the transmitter and receiver. Other models
take into account effects of multi-path fading and one of the most commonly used
is log-distance path loss model [15]:
PL = (1/d)α (3.15)
This model employe path loss exponent which is empirically measured under
different propagation scenarios. Using this model we can express receiving power
Pr at distance d from the transmitter [30]:
Pr = P0.(d0/d)α (3.16)
13
Table 3.2: Typical values of path exponent
Environment αFree-space 2
Urban area LOS 2,7 3,5Urban area no LOS 3 5
Indoor LOS 1,6 1,8Factories no LOS 2 3Buildings no LOS 4 6
Where P0 represents known received power at distance d0 from a transmitter and
is the path loss exponent. Pure theoretical model of wireless transmission, assumes
that all consumed energy is radiated into the air by a transmitter, and a receiver
does not spend any energy during a reception. Topologies of various types of
single-hop and multi-hop communication are presented in Fig3.3 [3].
If we assume that transmitter, in single-hop scenario, emits at such as power P1
which is just enough to be received by destination node, we can address this power
as receivers sensitivity threshold PM [31]:
PM = P1.(d0/d)α (3.17)
In case of the double-hop, triple-hop, quad-hop and n-hop necessary transmitting
powers P2, P3,P4,.., Pn will be [32]:
PM = P2.(d0/(d/2))α (3.18)
PM = P3.(d0/(d/3))α (3.19)
PM = P4.(d0/(d/4))α (3.20)
PM = Pn.(d0/(d/n))α (3.21)
If we equalize equations 2.3 2.7 we will get:
P1 = P2.2α = P3.3
α = P4.4α = Pn.n
α (3.22)
Over all transmitters power consumption used for single-hop P1H , double-hop
P2H , triple-hop P3H and n-hop PnH will be [33]:
P1H = P1 (3.23)
P2H = P2 + P2 = 2.(P1/α) (3.24)
14
P3H = P3 + P3 + P3 = 3.(P3/α) (3.25)
P4H = P4 + P4 + P4 + P4 = 4.(P4/α) (3.26)
PnH = n.(Pn/α) (3.27)
We can clearly see that for any value of the path loss exponent greater than one,
multi-hop transmission will be more energy efficient than single-hop transmission.
WSN nodes usually use transceivers, which operate in 2.45 GHz band, compliant
to IEEE 802.15.4 standard. This band has sixteen channels, each of them with
data rate of 250 kbps. It employs Direct Sequence Spread Spectrum (DSSS)
modulation in combination with Offset - Quadrate Phase Shift Keying (O-QPSK)
modulation. Radio transceiver has standard output power at 0 dBm and receivers
sensitivity threshold is at least of - 85 dBm.
15
START
Are Temperature Changes Detected in
Neighbour Nodes?
1:High RSSI_loss (A)
2:Medium RSSI_loss (B)
3:Low RSSI_loss (C)
1:RSSI_loss (A)
2:P_level (A)
3:Count N_A
4:N_A Desired
1:RSSI_loss (B)
2:P_level (B)
3:Count N_B
4:N_B Desired
1:RSSI_loss (C)
2:P_level (C)
3:Count N_C
4:N_C Desired
Define Threshold RSSI_loss
(A,B,C)
Broadcast
1:P_level_new(A,B,C)
2:P_save(A,B,C)
END
Yes
Keep Current
Transmitter Power
Level
No
RSSI_loss_Threshold (A,B,C)<=RSSI_loss (A,B,C)RSSI_loss_Threshold (A,B,C)>RSSI_loss (A,B,C)
N_current>=N_desired N_current<N_desired
RSSI_loss_new
(A,B,C)
=RSSI_loss_Thresh
old
RSSI_loss_new
(A,B,C)=RSSI_loss
(A,B,C)
Set the
parameters
(N,d,T)
Estimate
1:RSSI_loss
2:P_level
Figure 3.2: Flow Chart of Reference Node
16
Figure 3.3: Transmission distances for: (a) signle-hop, (b) double-hop, (c) triple-hop,(d) quad-hop.
17
Chapter 4
Results and Discussion
4.1 Simulation Results of EETS
In this section we analyze and describe results for required transmitter power
in WSNs for Frequency 2.45GHz, Environments (n=2,3,4), Coding techniques (
Convolutional, Reed-solomon) and also find bit error rate for given Eb/N0.
10 20 30 40 50 60 70 80 90 100−220
−200
−180
−160
−140
−120
−100
Distance(m)
Pt(
dB)
Figure 4.1: Required Transmitter power versus distance
Fig4.1 shows results between required transmitter power for free space path loss
model excluding multi-path fading effect at frequency 2.45GHz versus distance be-
tween sensor nodes (1-100)m. We see that required power increase with increasing
distance. After analyzing required transmitter power for given Eb/N0 8.3dB we
analyze transmitter power for estimated multi-path fading effect for given envi-
18
10 20 30 40 50 60 70 80 90 100−220
−200
−180
−160
−140
−120
−100
Distance(m)
Pt(
dB)
[email protected] with estimated [email protected] with average fading
Figure 4.2: Transmitter power versus distance for estimated fading
ronment and average that is given in range (4-10)dB [34].
We clearly see in Fig4.2 that for estimated power we need less power than standard
value that is given in range that saves energy and increase battery life time. Fig4.3
shows required power versus distance for signal without fading effect and with
fading effect we see that signal with fading required large power than signal without
fading effect. It means if we do not include that effect we transmit less power that
cause difficulty in signal detection.
As we have earlier analyze required power for free space model with and without
fading now we see that effect shown above required transmitter power versus
distance in Fig4.4 for various environments. We see that for n=2 (free space
) required power is minimum and for n=4 (large scattering) required power is
maximum and it increases with distance.
Fig4.5 shows that required power is maximum for Un-coded and less value for
coded system like RS, CC-Hard decision, CC-Soft decision and minimum for CC-
soft decision for frequency 2.45GHz. Fig4.6 shows required transmitter energy
that is transmitter power divide by data rate for coded and un-coded transmission
between sensor nodes. For un-coded sensor requires maximum energy and for CC-
soft decision it requires minimum energy. Fig4.7 shows bit error rate versus given
Eb/N0 for different coding techniques we see that by increasing Eb/N0 bit error
rate decreases slowly and for RS bit error rate decreases fast.
19
10 20 30 40 50 60 70 80 90 100−220
−210
−200
−190
−180
−170
−160
−150
−140
−130
−120
Distance(m)
Pt(
dB)
Pt(Without fading)Pt(With fading)
Figure 4.3: Transmitter power versus distance with and without fading
4.2 Simulation Results of EAST
In this section we describe simulation results of our propose technique for power
efficient transmission in WSNs compared with classical approach used for trans-
mission of data. In Fig4.8 we have shown values of meteorological temperature
for one round that each sensor node have sensed in WSN. Let suppose we have
100 nodes in 100*100 m2 square region and temperature have values in range (-
10 - 53)Co [8] for given meteorological condition of Pakistan. Reference node is
placed at edge of this region. In figure shown earlier temperature variation shown
on y-axis and corresponding nodes on x-axis. Each sensor node placed at different
location randomly in given area and we clearly see variation of temperature for
different nodes in WSN.
Different values of temperature for each sensor node based on meteorological con-
dition help to estimate RSSIloss(dBm) that is transmitter power loss. Fig4.9
shows transmission power loss due to temperature variation in any environment
using the relationship between RSSIloss(dBm) and temperature (Co) given by
Bannister et al. RSSIloss(dBm) on y-axis indicates transmission power loss for
each sensor node. RSSIloss(dBm) high means that sensor node placed in region
where temperature is high so link not have good quality. For temperature (-10 -
53)Co RSSIloss(dBm) have value in range (-6dBm) - (5dBm).
From figure shown earlier it is also clear that link quality and RSSIloss have inverse
relation, when temperature is high RSSIloss has high value means low quality link
and vise versa. As we have earlier mentioned link quality and RSSIloss have
20
10 20 30 40 50 60 70 80 90 100−250
−200
−150
−100
−50
0
50
100
Distance(m)
Pt(
dB)
[email protected][email protected][email protected]
Figure 4.4: Transmitter power versus distance for various environments
Table 4.1: Simulation Parameters
Rounds 1200Temperature (-10-53) C0
Distance (1-100) mNodes 100Regions A,B,C
η 0.0029SNR 0.020 (dB)
Bandwidth 83.5MHzFrequency 2.45GHz
RNF 5dBT(Absolute) 300k
Eb/No 8.3 dB
inverse relation that is for high temperature link quality is not good and for low
temperature link quality is good. After estimating RSSIloss for each node in WSN
we compute corresponding transmitter power level to compensate RSSIloss. Plevel
assigned to each node on basis of nodes estimated RSSIloss. Fig4.10 shows range
of power levels on y-axis for given RSSIloss that is between (20- 47) dBm and
also variation of required power level for sensor node with changing temperature
that is at low temperature required Plevel is low and for high temperature required
Plevel is high.
As we have earlier estimated RSSIloss for each sensor node on the basis of given
meteorological temperature that help to estimate required power level to compen-
sate transmission power loss. That power level only help to compensate RSSIloss
21
10 20 30 40 50 60 70 80 90 100−220
−210
−200
−190
−180
−170
−160
−150
−140
−130
−120
Distance(m)
Pt(
dB)
UncodedRS (255,239)CC HARD DECISIONCC SOFT DECISION
Figure 4.5: Transmitter power versus distance for various channel coding techniques
Table 4.2: Estimated Parameters
Number of Nodes (A,B,C) 46,30,24Desired Neighbors 41,25,19
Number of Nodes after 1200 Rounds (A,B,C) 41,22,17Threshold power level (A,B,C) 43.24,31.77,22.21 dBmNodes above threshold (A,B,C) 23,11,8Nodes below threshold (A,B,C) 18,11,9
PRR (A,B,C) (80-98),(70-96),(63-97) %Threshold RSSIloss ( A,B,C) 3.78,-0.61,-5.17 dBm
due to temperature variation. To compensate path loss due to distance between
each sensor node in WSN free space model help to estimate actual required trans-
mitter power. After addition of required power level due to temperature variation
and distance, we estimate actual required transmitter power between each sen-
sor node. Fig4.11 shows required transmitter power including both transmission
power loss due to temperature variation and free space path loss for different
nodes. We clearly see from figure that Pt lies between (-115 - 45)dBm and most
of times it is above -100dBm .
Table4.1 shows simulation parameters required for estimation of required power
level both for temperature variation and free space path loss model. For estimation
of required power level due to temperature variation we need values of temperature
and for free space path loss model number of nodes, distance between each node,
required Eb/No depends upon SNR, η, f and (RNF ).
As we have chosen 1200 rounds for our analysis each round starts when tempera-
22
10 20 30 40 50 60 70 80 90 1000
0.5
1
1.5
2
2.5
3
3.5x 10
−18
Distance(m)
Et(
J/bi
t)
UncodedRS(255,239)CC Hard DecisionCC Soft Decision
Figure 4.6: Transmitter energy versus distance for different coding techniques
ture change detected also we have divided network into three regions (A,B,C) for
analysis of our propose technique. Table4.2 below shows estimated parameters like
number of nodes in each region based on RSSIloss, Threshold RSSIloss for each
region, nodes above and below threshold in each region, packet reception ratio for
each region based on current number of neighbor nodes and desired number of
nodes and threshold power level for each region after 1200 rounds. From sensor
nodes sensed temperature we have estimated RSSIloss that describes transmission
power loss due to temperature variation. After that we have assigned RSSIloss to
each node. In our approach we have divided network into three regions on basis of
Threshold RSSIloss and count numbers of nodes in each region. Nodes with high
RSSIloss in region (A), medium RSSIloss in (B) and low RSSIloss in (C).
After estimating RSSIloss for nodes of each region we have estimated required
Plevel for nodes of each region that we clearly see in Fig4.12, in region A Plevel lies
between (40-45)dBm, for region B (30-35) dBm and for region C (20-25)dBm.
It means that for region A required power level high then both other region that
also shows that for that region temperature and RSSIloss is large. For region B
required power level is between both region A and C and for C region required
power level is less then both other two regions. We have earlier seen in Fig4.13
power level for each region assigned using classical approach. After applying our
propose technique we see what power level required for each region. We clearly see
difference between Plevel as shown in Fig4.13, that required power level decrease for
each region and for region A it decreases maximum. Fig4.14, 4.15, 4.16 respectively
shows required Power level save for region A,B and C respectively after implanting
23
1 2 3 4 5 6 7 8 9 100
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
Eb/No (dB)
BE
R
CC HARDCC SOFTRS(255,239)
Figure 4.7: Bit error rate versus given Eb/No
propose technique. We save power up to 2.3dBm for region A, 1.7dBm for B and
1.5dBm for C.
Earlier we have shown results of power save in different regions for fixed reference
node location. Now we will see that how by changing reference node location
power level save effected in different regions. When we place reference node at
edge of area we see that most of times power not saved because most of nodes not
into coverage of reference node. But when we place reference node at center of
region most of nodes are into coverage of reference node and we save maximum
power and power level save not go to zero. Similarly when reference node at corner
of area some nodes not into coverage of reference node and we save less power than
center location but more power save than edge location. In Fig4.17 we have shown
power level save for region A in different reference node location scenarios (Edge,
Center, Corner) by taking mean of power save for 1200 rounds 10 times. We
can clearly see that in figure that power save is maximum for center location and
minimum for edge location. We can also see that difference between power save
for edge and center location is 0.6dBm. For Corner location it is between both
locations. Similar behaviour we can see for both other regions B and C in Fig4.18
and Fig4.19 respectively. For region B difference between center and edge location
is 0.25dBm and for region C the difference is 0.3dBm
24
10 20 30 40 50 60 70 80 90 100−10
0
10
20
30
40
50
60
Nodes (N)
Tem
pera
ture
(C
o)
Temperature (Co)
Figure 4.8: Temperature for different Nodes
4.3 Simulation Results of MEAST
Fig4.20 shows result for required transmitter power level for region A using multi-
hop in different environment. We can clearly see in figure that when number
of hops increased on x-axis then transmitter power level decrease.That shows
that multi-hop communication certainly reduces the transmitter power for each
node in region.This figure also shows transmitter power level for different environ-
ments.Similarly Fig4.21 shows result for required transmitter power level for region
B using multi-hop in different environment. We can clearly see in that figure that
when number of hops increased on x-axis then transmitter power level decrease.
This figure also shows transmitter power level for different environments.Similarly
Fig4.22 shows result for required transmitter power level for region C using multi-
hop in different environment. We can clearly see in that figure that when number
of hops increased on x-axis then transmitter power level decrease. This figure also
shows transmitter power level for different environments.
25
10 20 30 40 50 60 70 80 90 100−8
−6
−4
−2
0
2
4
6
Nodes (N)
RS
SI−
loss
(dB
m)
RSSI−loss (dBm)
Figure 4.9: RSSI-loss for different Nodes
10 20 30 40 50 60 70 80 90 10015
20
25
30
35
40
45
50
Nodes (N)
Pow
er le
vel (
dBm
)
Power level (dBm)
Figure 4.10: Power level for different Nodes
26
10 20 30 40 50 60 70 80 90 100−120
−110
−100
−90
−80
−70
−60
−50
−40
Nodes (N)
Pt (
dBm
)
Pt (dBm)
Figure 4.11: Transmitter Power for different Nodes
100 200 300 400 500 600 700 800 900 1000 1100 120020
25
30
35
40
45
Rounds
Pow
er le
vel(A
,B,C
) dB
m
Power level(A,B,C) dBm
Figure 4.12: Power level for different regions
27
100 200 300 400 500 600 700 800 900 1000 1100 120020
25
30
35
40
45
Rounds
Pow
er le
vel n
ew(A
,B,C
) dB
m
Power level new(A,B,C) dBm
Figure 4.13: Power level using EAST for different regions
100 200 300 400 500 600 700 800 900 1000 1100 12000
0.5
1
1.5
2
2.5
Rounds
Pow
er le
vel s
ave(
A)
dBm
Power level save(A) dBm
Figure 4.14: Power level save for region A
28
100 200 300 400 500 600 700 800 900 1000 1100 12000
0.5
1
1.5
2
2.5
Rounds
Pow
er le
vel s
ave(
B)
dBm
Power level save(B) dBm
Figure 4.15: Power level save for region B
100 200 300 400 500 600 700 800 900 1000 1100 12000
0.2
0.4
0.6
0.8
1
1.2
1.4
Rounds
Pow
er le
vel s
ave(
C)
dBm
Power level save(C) dBm
Figure 4.16: Power level save for region C
29
1 2 3 4 5 6 7 8 9 100.9
1
1.1
1.2
1.3
1.4
1.5
1.6
Rounds
Ple
vels
ave(
A)
dBm
EdgeCenterCorner
Figure 4.17: Power level save in region A for different Reference Node Locations
1 2 3 4 5 6 7 8 9 100.7
0.8
0.9
1
1.1
1.2
1.3
1.4
Rounds
Ple
vels
ave(
B)
dBm
EdgeCenterCorner
Figure 4.18: Power level save in region B for different Reference Node Locations
30
1 2 3 4 5 6 7 8 9 100.65
0.7
0.75
0.8
0.85
0.9
0.95
1
1.05
1.1
1.15
Rounds
Ple
vels
ave(
C)
dBm
EdgeCenterCorner
Figure 4.19: Power level save in region C for different Reference Node Locations
1 2 3 4 5 6 7 8 9 100
5
10
15
20
25
30
35
40
45
Number of Hops
Ple
vel (
A)
dBm
n=2n=3n=4
Figure 4.20: Power level for region A in different environments
31
1 2 3 4 5 6 7 8 9 100
5
10
15
20
25
30
35
Number of Hops
Ple
vel (
B)
dBm
n=2n=3n=4
Figure 4.21: Power level for region B in different environments
1 2 3 4 5 6 7 8 9 100
5
10
15
20
25
Number of Hops
Ple
vel (
C)
dBm
n=2n=3n=4
Figure 4.22: Power level for region C in different environments
32
Chapter 5
Conclusion
In this thesis, I have presented my propose technique EAST to study the ef-
fect of temperature on wireless link quality. That shows temperature is one of
most important factors impacting link quality. Relationship between RSSIloss
and temperature variation has been analyzed for propose adaptive transmission
power control scheme. This scheme uses open-loop control to compensate changes
of link quality due to temperature variation. Combining open-loop temperature
aware compensation and close-loop feedback control significantly minimize over-
head of propose transmission power control scheme in WSN. I further extended
my scheme by dividing network into three regions on basis of Threshold RSSIloss
and assign power level to each node in three regions on basis of current number
of nodes and desired number of nodes. Which help to adapt transmitter power
according to link quality variation and also increase network lifetime.
In free space line-of-sight environment, ECC is not very energy efficient for fre-
quencies below 2GHz. ECC can be practical for WSN placed between buildings,
especially when implemented with analog decoders. For indoor environments ECC
is energy-efficient at high frequencies. ECC is not always a practical solution for
increasing link reliability, especially when implemented with analog decoders. So
to increase link reliability in presence of existing error control techniques and de-
coder complexity we estimate transmitter power for given environment to increase
energy efficiency. Otherwise we have to transmit more power most of time that
limit battery life time of sensor node. Required transmitter power for sensor nodes
in given environment for different coding techniques like Reed Solomon and Con-
volution codes energy efficiency and bit error rate has been analyzed for different
Eb/N0.
33
Bibliography
[1] I. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, “Wireless sensor
networks: a survey,” Computer networks, vol. 38, no. 4, pp. 393–422, 2002.
[2] V. Raghunathan, C. Schurgers, S. Park, and M. Srivastava, “Energy-aware
wireless microsensor networks,” Signal Processing Magazine, IEEE, vol. 19,
no. 2, pp. 40–50, 2002.
[3] F. Lavratti, A. Ceratti, D. Prestes, A. Pinto, L. Bolzani, F. Vargas, C. Mon-
tez, F. Hernandez, E. Gatti, and C. Silva, “A transmission power self-
optimization technique for wireless sensor networks,” ISRN Communications
and Networking, vol. 2012, p. 1, 2012.
[4] V. Douros and G. Polyzos, “Review of some fundamental approaches for
power control in wireless networks,” Computer Communications, vol. 34,
no. 13, pp. 1580–1592, 2011.
[5] J. Lee and K. Chung, “An efficient transmission power control scheme for
temperature variation in wireless sensor networks,” Sensors, vol. 11, no. 3,
pp. 3078–3093, 2011.
[6] X. Cui, X. Zhang, and Y. Shang, “Energy-saving strategies of wireless sensor
networks,” in Microwave, Antenna, Propagation and EMC Technologies for
Wireless Communications, 2007 International Symposium on, pp. 178–181,
IEEE, 2007.
[7] A. Eswaran, A. Rowe, and R. Rajkumar, “Nano-rk: an energy-aware resource-
centric rtos for sensor networks,” in Real-Time Systems Symposium, 2005.
RTSS 2005. 26th IEEE International, pp. 10–pp, IEEE, 2005.
[8] M. Meghji and D. Habibi, “Transmission power control in multihop wireless
sensor networks,” in Ubiquitous and Future Networks (ICUFN), 2011 Third
International Conference on, pp. 25–30, IEEE, 2011.
[9] K. Bannister, G. Giorgetti, and S. Gupta, “Wireless sensor networking for hot
applications: Effects of temperature on signal strength, data collection and
34
localization,” in Proceedings of the fifth Workshop on Embedded Networked
Sensors (HotEmNets 08), Citeseer, 2008.
[10] S. Cheema, G. Rasul, and D. Kazmi, “Evaluation of projected minimum
temperatures for northern pakistan,” 2010.
[11] L. Howard Sheryl, S. Christian, I. Kris, and I. Kris, “Error control cod-
ing in low-power wireless sensor networks: When is ecc energy-efficient?,”
EURASIP Journal on Wireless Communications and Networking, vol. 2006.
[12] M. Pellenz, R. Souza, and M. Fonseca, “Error control coding in wireless sensor
networks,” Telecommunication Systems, vol. 44, no. 1, pp. 61–68, 2010.
[13] E. Sanchez, F. Gandino, B. Montrucchio, and M. Rebaudengo, “Increasing
effective radiated power in wireless sensor networks with channel coding tech-
niques,” in Electromagnetics in Advanced Applications, 2007. ICEAA 2007.
International Conference on, pp. 403–406, IEEE, 2007.
[14] P. Data, “Prediction methods required for the design of earth-space telecom-
munication systems,” Recommendation P. 618-8, ITU-R P Sers, 2003.
[15] D. Anand, H. Chandrakanth, and M. Giriprasad, “An efficient energy, cover-
age and connectivity (ec2) algorithm for wireless sensor networks,” Interna-
tional Journal of Computer Applications (0975–8887) Volume, 2012.
[16] A. Nandi and S. Kundu, “On energy level performance of adaptive power
based wsn in presence of fading,”
[17] S. Kasnavi, S. Kilambi, B. Crowley, K. Iniewski, and B. Kaminska, “Ap-
plication of error control codes (ecc) in ultra-low power rf transceivers,” in
Architecture, Circuits and Implementtation of SOCs, 2005. DCAS’05. Pro-
ceedings of the 2005 IEEE Dallas/CAS Workshop:, pp. 195–198, IEEE, 2005.
[18] Q. Wang, M. Hempstead, and W. Yang, “A realistic power consumption
model for wireless sensor network devices,” in Sensor and Ad Hoc Commu-
nications and Networks, 2006. SECON’06. 2006 3rd Annual IEEE Commu-
nications Society on, vol. 1, pp. 286–295, IEEE, 2006.
[19] B. Sharma and M. Singh, “New era of power efficent back-off algorithm
trend in wireless sensor network,” INTERNATIONAL JOURNAL OF ELEC-
TRONICS & DATA COMMUNICATION, vol. 1, no. 1, pp. 5–7, 2012.
[20] N. Sadeghi, S. Howard, S. Kasnavi, K. Iniewski, V. Gaudet, and C. Schlegel,
“Analysis of error control code use in ultra-low-power wireless sensor net-
35
works,” in Circuits and Systems, 2006. ISCAS 2006. Proceedings. 2006 IEEE
International Symposium on, pp. 4–pp, IEEE, 2006.
[21] B. Liu, B. Otis, S. Challa, P. Axon, C. Chou, and S. Jha, “On the fading and
shadowing effects for wireless sensor networks,” in Mobile Adhoc and Sensor
Systems (MASS), 2006 IEEE International Conference on, pp. 51–60, IEEE,
2006.
[22] C. Phan, Y. Park, H. Choi, J. Cho, and J. Kim, “An energy-efficient trans-
mission strategy for wireless sensor networks,” Consumer Electronics, IEEE
Transactions on, vol. 56, no. 2, pp. 597–605, 2010.
[23] N. Xia, M. Tang, J. Jiang, D. Li, and H. Qian, “Energy efficient data trans-
mission mechanism in wireless sensor networks,” in Computer Science and
Computational Technology, 2008. ISCSCT’08. International Symposium on,
vol. 1, pp. 216–219, IEEE, 2008.
[24] M. Islam, “Selection of error control/correction codes in wireless sensor net-
work,” in Electrical and Computer Engineering (ICECE), 2010 International
Conference on, pp. 674–677, IEEE, 2010.
[25] A. Nandi and S. Kundu, “Optimal transmit power in wireless sensor networks
in a multipath rician fading channel,” in India Conference (INDICON), 2010
Annual IEEE, pp. 1–5, IEEE, 2010.
[26] D. Son, B. Krishnamachari, and J. Heidemann, “Experimental study of the
effects of transmission power control and blacklisting in wireless sensor net-
works,” in Sensor and Ad Hoc Communications and Networks, 2004. IEEE
SECON 2004. 2004 First Annual IEEE Communications Society Conference
on, pp. 289–298, IEEE, 2004.
[27] K. Souccar, Transmission power control for wireless sensors networks. PhD
thesis, University of South Florida, 2006.
[28] Y. Yu, C. Song, M. Liu, and H. Gong, “Energy-efficient algorithm for sensor
networks with non-uniform maximum transmission range,” Sensors, vol. 11,
no. 6, pp. 6203–6213, 2011.
[29] H. She, Z. Lu, A. Jantsch, D. Zhou, and L. Zheng, “Analytical evaluation of
retransmission schemes in wireless sensor networks,” in Vehicular Technology
Conference, 2009. VTC Spring 2009. IEEE 69th, pp. 1–5, IEEE, 2009.
[30] V. Sachan, S. Imam, and M. Beg, “Energy-efficient communication meth-
ods in wireless sensor networks: A critical review,” International Journal of
Computer Applications, vol. 39, no. 17, 2012.
36
[31] Y. Yu and V. Prasanna, “Energy-balanced multi-hop packet transmission in
wireless sensor networks,” in Global Telecommunications Conference, 2003.
GLOBECOM’03. IEEE, vol. 1, pp. 480–486, IEEE, 2003.
[32] J. Wang, J. Cho, and S. Lee, “Experimental study of energy consumption in
direct transmission and multi-hop transmission for wireless sensor networks,”
in Networking, Sensing and Control, 2008. ICNSC 2008. IEEE International
Conference on, pp. 1179–1184, IEEE, 2008.
[33] J. Li and G. AlRegib, “Network lifetime maximization for estimation in mul-
tihop wireless sensor networks,” Signal Processing, IEEE Transactions on,
vol. 57, no. 7, pp. 2456–2466, 2009.
[34] Y. Huang and Y. Hua, “Energy cost for estimation in multihop wireless sensor
networks,” in Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE
International Conference on, pp. 2586–2589, IEEE, 2010.
37
Appendix
1. Wireless Sensor Network
WSNconsists of spatially distributedautonomoussensorstomonitor physical or en-
vironmental conditions such astemperature,pressure, etc and to cooperatively pass
their data through the network to a main location. The development of WSNs
was motivated by military applications such as battlefield surveillance; today such
networks are used in many industrial and consumer applications, such as industrial
process monitoring and control, machine health monitoring, and so on.
2. Control Packets
A packet consists of two kinds of data: control information and user data (also
known aspayload). The control information provides data the network needs to de-
liver the user data, for example: source and destination addresses, error detection
codes like checksums, and sequencing information. Typically, control information
is found in packetheadersand trailers.
3. IEEE802.15.4
IEEE802.15.4is a standard which specifies thephysical layerandmedia access con-
trolfor low-rate wirelesspersonal area networks(LR −WPANs). It is maintained
by theIEEE 802.15working group. It is the basis for theZigBee, andMiWispecifications,
each of which further extends the standard by developing the upper layerswhich
are not defined in IEEE 802.15.4.
4. Transmitter Power Loss
Transmitter power loss (RSSIloss) defined as loss in transmitter power that cause
link quality degradation due to temperature variation .
5. Reference Node
Reference node rn periodically broadcasts a beacon message to neighbor nodes nn.
6. Beacon Message
A message frame sent repeatedly by an reference node rn indicating a temperature
change detected in network .
7. Neighbour Node
Neighbor nodes nn hear a beacon message from a reference node rn. The nodes
exceeding the threshold RSSIloss are regarded as the neighbor nodes with reliable
links.
8. Local Mean Algorithm
38
In Local Mean Algorithm (LMA), a reference node broadcasts the LifeMsg mes-
sage. neighbor nodes transmit the LifeAckMsg after they receive LifeMsg.
Reference node count the number of LifeAckMsgs and the transmission power
is controlled by maintaining appropriate connectivity.
9. Dynamic Transmitter Power Control
Dynamic Transmission Power Control (DTPC) uses the RSSIloss to estimate
transmitter power level. The nodes exceeding the threshold RSSIloss are regarded
as the neighbor nodes with reliable links.
10. Transmitter Power Level
Transmitter power level Plevel helps to compensate loss due to corresponding tem-
perature variation.
11. Pcket Reception Ratio
Packet reception ratio (PRR) is defined as the ratio of desired number of nodes
nd(t) minus current number of nodes nc(t) to desired number of neighbor nodes.
12. Spectral Efficiency
Spectral efficiency (eta) is defined as ratio of data rate (R) to bandwidth (B).
13. Signal to Noise Ratio
Signal to noise ratio (SNR) is defined as required signal power to noise power.
14. Receiver Noise Figure
Receiver noise figure (RNF ) is defined as input noise at receiver to output noise
at receiver.
15. Path Loss
Path loss (PL) is defined as difference between received power (Pr) minus trans-
mitted power (Pt).
16. Bandwidth
Bandwidthis the difference between the upper and lower frequencies in a continu-
ous set of frequencies. It is typically measured inhertz, and may sometimes refer
topass-band bandwidth.
17. Error Control Coding (ECC)
Method in which redundancies introduced into data to be transmitted. Transmit-
ter enables receiver to detect or correct some errors.
39
18. Critical Distance
Distance at which decoder energy consumption becomes equal to transmit energy
saving.
19. Coding Gain
Ability of encoder to provide better BER over noisy channel for same signal to
noise ratio as compared to an un-coded systems.
20. Bit Error Rate
Ratio of bit errors in received bits of data to total transferred bits in unit time is
called bit error rate.
21. Automatic Repeat Request
Error correction method an ack is sent after successful reception of data, if no ack
received in specific time then data is retransmitted.
22. Forward Error Correction
Redundant bits are sent with data by using ECC technique. Redundancy al-
lows receiver to correct limited number of errors and correction is made without
retransmission.
23. Required Transmit Power
Minimum transmit power at which data can be successfully transmitted without
error.
24. Noise Power Spectral Density
Noise power per unit bandwidth.
25. Spectral Efficiency
Information that can be transmitted over given bandwidth.
40