research article fuzzy weight cluster-based routing...

Research ArticleFuzzy Weight Cluster-Based Routing Algorithm forWireless Sensor Networks

Teng Gao12 Jin-Yan Song3 Jin-Hua Ding1 and De-Quan Wang1

1School of Mechanical Engineering amp Automation Dalian Polytechnic University Dalian 116034 China2School of Control Science and Engineering Dalian University of Technology Dalian 116024 China3School of Information Engineering Dalian Ocean University Dalian 116023 China

Correspondence should be addressed to Teng Gao gaotengdlpueducn

Received 10 April 2015 Revised 28 July 2015 Accepted 26 August 2015

Academic Editor Onur Toker

Copyright copy 2015 Teng Gao et alThis is an open access article distributed under theCreative CommonsAttribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Cluster-based protocol is a kind of important routing inwireless sensor networksHowever due to the uneven distribution of clusterheads in classical clustering algorithm some nodes may run out of energy too early which is not suitable for large-scale wirelesssensor networks In this paper a distributed clustering algorithm based on fuzzy weighted attributes is put forward to ensure bothenergy efficiency and extensibility On the premise of a comprehensive consideration of all attributes the corresponding weightof each parameter is assigned by using the direct method of fuzzy engineering theory Then each node works out property valueThese property values will be mapped to the time axis and be triggered by a timer to broadcast cluster headers At the same timethe radio coverage method is adopted in order to avoid collisions and to ensure the symmetrical distribution of cluster headsThe aggregated data are forwarded to the sink node in the form of multihop The simulation results demonstrate that clusteringalgorithm based on fuzzy weighted attributes has a longer life expectancy and better extensibility than LEACH-like algorithms

1 Introduction

In recent years applications of sensor networks have evolvedin many areas due to their large applicability and develop-ment possibilities especially in the wireless sensor networks(WSNs) area [1] A wireless sensor network consists of a largenumber of light-weight sensor nodes having limited batterylife computational capabilities storage and bandwidth [2]The potential applications of sensor networks are highlyvaried such as environmental monitoring target trackingbattle field surveillance monitoring the enemy territorydetection of attacks and security etiquette [3] An importantaspect of such networks is that the nodes are unattendedresource-constrained their energy cannot be replenishedand network topology is unknown [4] The node which lostenergy may cause the malfunction of the entire networkTherefore the research on WSNs has mainly been focusedon saving the limited energy and extending the life time ofwireless sensor networks Researchers have developed manytheories to save energy from almost every aspect but we haveour sight on routing protocol

An efficient routing protocol is the one which consumesminimum energy and provides large coverage area [5] Basedon the logical structure the routing protocol is divided intotwo categories The first category is flat routing in whichall nodes in the network are coequal and there are nospecial nodes The advantage of this type of protocols is theirrobustness The other category is hierarchical-based routingOne of the most classical paradigms of hierarchical-basedrouting is the clustering in which cluster is an infrastructureand nodes play different roles In a clustering architecturecluster head nodes can be used to process and send theinformation to the sink node while member nodes can beused to perform the sensing in the proximity of the targetand transmit the information to corresponding cluster headClustering provides an efficient way of saving energy withina cluster and outside cluster and inside a wireless sensornetwork The cluster head acts as a bridge between othersensor nodes and sink node and sometimes between onecluster head and other cluster head in multihop cases [6]This means that creation of clusters and assigning specialtasks to cluster heads can greatly contribute to overall system

Hindawi Publishing CorporationJournal of Control Science and EngineeringVolume 2015 Article ID 583092 12 pageshttpdxdoiorg1011552015583092

2 Journal of Control Science and Engineering

scalability lifetime and energy efficiency [7] Thus cluster-based routing takes great advantage over the plat-based oneat above performances However the disadvantage of cluster-based routing is that the cluster head is so vital that it becomesthe bottleneck of the entire network Therefore the selectionof the cluster head will influence the performance of theentire network The existing clustering algorithms differ onthe criterium for the selection of the cluster heads Accordingto the current research findings the selection method of thecluster head can be divided into several categories below

The classical routing protocols that are based on k-meansclustering such as LEACH [8] TEEN [9] and APTEEN [10]select cluster head based on a random acquired value Ifthis value is less than a certain threshold the nodes willbe the cluster head Whereas because of the randomicityduring the selection the selected cluster head is prone tobe distributed improperly and unevenly this could cause theuneven distribution of the traffic flow in different cluster headnodes One of the direct consequences is that some clusterheads exhaust energy at the mean time the performance ofthe entire network is affected

Some distributed routing protocols based on a certainattribute are proposed in DCHS [11] HCDA (the Highest-ConnectivityDegreeAlgorithm) [12] andACMWN[13]Theattributes that can determine cluster head selection includeresidual energy neighbors number the cost that communi-cate in intracluster and the distance between the node andthe sink node and ID Because only one attribute is taken intoaccount in these protocols the selected cluster head cannotbe the most suitable node Although the rationality of thecluster head selection is improved to a certain extent certainproblems such as the unevenly distributed cluster head andthe imbalance load remain unsolved

Multiattribute cluster head selection protocol such asHEED [14] and WCA [15] which use several attribute todetermine the cluster head are greatly favored due to the con-sideration of various factors The advantage of multiattributecluster head selection is that a better partition of clustercan be obtained The two protocols both adopt successivescreeningmethod to determine the cluster head bywhich thefinite iteration must be implemented The major drawbackof the former is that distributed algorithm makes each nodeunaware of global information so that some nodes may notjoin any clusters while the latter need to iterate many times ifmany attributes are used to gain a better performance whichwill increase time complexity and consume more energy

WCA-LEACH [16] MWBC [17] WCA-GSEN [18] andAOW-LEACH [19] combine multifactors such as residualenergy communication cost and neighbor nodes number inorder to avoid the randomicity in the cluster head selectionof LEACH However all the algorithms above determineweight of each factor using trial and error method which willinfluence the performance of the whole protocol

From the analysis and comparison mentioned abovemultiattribute cluster head selection can obtain the unparal-leled rationality in partition of cluster therefore we considerthe residual energy neighbors number the cost that commu-nicate in intracluster and the distance between the node andthe sink node as the attribute to propose a new clustering

routing algorithm using fuzzy weight multiattribute (CFWA)to determine cluster head selection by which the energycan be saved and the lifetime of the whole network will beextended

2 Conform the Weight of Attribute

To save energy and balance load the residual energy is themost crucial factor of the attributes during the process ofcluster head selection The cost that is used in the com-munication in intracluster and the neighbors number alsoinfluence cluster head selection Nevertheless the distancebetween the node and the sink node will not be considereddue to the fact that uniform distribution of the cluster head isrequired by energy-efficient cluster-based algorithm In thispaper the direct method [20] based on the abutting objectrelative membership degree in engineering fuzzy theory andintelligence decision-making is adopted in order to confirmthe proportion of each attribute that is hold during clusterhead selection

Definition 1 Compare the member119874119896 with another member119874119897 on duality about weightiness in the object set119874 When119874119896is more important than 119874119897

05 lt 120573119896119897 ⩽ 1 (1)

when 119874119897 is more important than 119874119896

0 ⩽ 120573119896119897 lt 05

120573119896119897 = 1 minus 120573119897119896

(2)

when 119874119896 has the same importance as 119874119897

120573119896119897 = 05

especially 120573119896119896 = 05

(3)

where120573119896119897 is named relatively weightiness fuzzy value betweenthe object 119874119897 and 119874119896 Particularly if the object sequencingabout weightiness is 1198741 ≺ 1198742 ≺ sdot sdot sdot ≺ 119874119898 12057311989611198961+1 (1198961 =1 2 119898 minus 1) is defined as the abutting object relativelyweightiness fuzzy value

Assumption 2 In the available attributes it is assumed thatresidual energy (119864119903) is more significant than the cost thatcommunicate in intracluster (Cost) and the latter is moreimportant than neighbors number (Deg)

That is residual energy has the unexampled importancethan the cost that communicate in intracluster while the latteris more important than neighbors number ratherish Therelevant fuzzy value that 120573119864119903Cost is 1 and 120573CostDeg is 055 canbe found out based on Table 1 [21]

Based on the assumption about relative significance fuzzyscale value provided that the object 119864119903 is more importantthan Cost 120573119864119903Cost is the corresponding significance degreewhen just comparing objects 119864119903 and Cost of which bench-mark is 119864119903 the more important one between these twoobjects Because 120573119864119903119864119903 which is the fuzzy scale value that

Journal of Control Science and Engineering 3

Table 1 Relationships between mood operator and fuzzy value

Mood operator Fuzzy scale value Memberships valueSimilar 05 1Ratherish 055 0905Slightly 06 0667Relatively 065 0538Obviously 07 0429Markedly 075 0333Quite 08 025Very 085 0176Extremely 09 0111Violently 095 0053Incomparable 1 0

the object 119864119903 compares to itself is 05 if the only two objects119864119903 and Cost are still compared the degree that 119864119903 belongsto significance is 120573119864119903 = 1 and the one of Cost is 120573Cost =

15 minus 120573119864119903Cost Therefore the relationship of the significancedegree between 119864119903 and Cost is

120573119864119903

120573Cost=

1

15 minus 120573119864119903Cost (4)

The nonnormalization weight may be figured out

1205961015840

119864119903= 1

1205961015840

Cost = 1205961015840

119864119903(15 minus 1) = 05

1205961015840

Deg = 1205961015840

Cost (15 minus 055) = 0475

(5)

The object weight vector that obtained after normaliza-tion and reverting suffix is

120596 = (120596119864119903 120596Cost 120596Deg) = (05063 02532 02405) (6)

On account of the diffidence in the unit of each attributenormalization procedure will be implemented

The normalization expression of residual energy is119864119903119864max in which 119864max is the original energy of each nodeA proportional function relationship exists between residualenergy and cluster head selection that is the node whoseresidual energy is higher has more chances to be a clusterhead

The costs that communicate in intracluster which isobtained by calculating received signal strength information(RSSI) is normalized as RSSIaveTSSI where TSSI is thetransmission signal strength value which will be the samein the broadcasting phase of each node RSSIave denotes theaverage strength value of all the wireless signals that havebeen received The bigger the value is the lower the cost isRSSIave is also proportional to the probability that the nodecan be selected as the cluster head

According to the conclusion that is drawn byHeinzelmanet al [22] the relationship of cluster number and energyconsumption in the scene of 100 nodes is showed in Figure 1from which the optimal nodes number in a cluster can

20

25

30

35

40

45

50

55

60

65

Aver

age e

nerg

y di

ssip

atio

n pe

r rou

nd (J

)

2 3 4 5 6 7 8 9 10 111Number of clusters

Figure 1 The relationship of cluster number and energy consump-tion in the scene of 100 nodes in LEACH

be deduced in other words the optimal neighbor numbercan be confirmed The normalization function 119865(Deg) thatdenotes the neighbor number and the energy expenditurerelationship is fit based on Figure 1

119865 (Deg) = 6479 times 1199092minus 2266119909 + 3001

119899 times 119864max times (1199092 minus 1017119909 + 2434)

(7)

Here 119899 is the total number of nodesThe absolute attribute degree value of each node based on

OWA operator can be calculated out by the following objectfunction

1198651 = 120596119864119903 sdot (119864119903

119864max) + 120596Cost sdot (

RSSIaveTSSI

) + 120596Deg

sdot 119865 (Deg) (8)

In the same manner the function

1198652 = (119864119903

119864max)

120596119864119903

+ (RSSIaveTSSI

)

120596Cost

+ 119865 (Deg)120596Deg (9)

is the absolute attribute degree value based on GOWA plusoperator while the expression

1198653 = (119864119903

119864max)

120596119864119903

sdot (RSSIaveTSSI

)

120596Cost

sdot 119865 (Deg)120596Deg (10)

calculates the absolute attribute degree value based onGOWAmultiplication operator

Obviously 120596119864119903 + 120596Cost + 120596Deg = 1 (11)

3 System Module

31 Network Module

(a) All sensor nodes cannot move after being deployedand each node has a unique ID


(b) There is the only one sink node which lies outside thenetwork

(c) All sensor nodes are homogeneous with no GPSequipment on it All nodes are time synchrony

(d) Each node has the ability to aggregate data as aresult several data packages can be compressed as onepackage

(e) If the node knows the transmission power it cancalculate out the approximate distance between thetransmitter and receiver based on the RSSI

RSSI = 119860 minus 10119899 log 10 (119889) (12)

where 119889 represents the distance 119860 is the RSSI valuewhen transmitter and receiver are 1m apart 119899 is theenvironmental factor

(f) The battery that cannot be supplied is themain energysupply of the node However the node is able to adjusttransmission power freely to save energy based on thedistance from the receiver

(g) The energy of the sink node is infinite(h) The bidirectional channel is defined through the

whole network

32 Wireless Channel Module The same wireless channelmodule is put to use in LEACH [8] and this paper whichis composed of free space module and two-ray groundmoduleThe boundary distance 1198890 is used to differentiate theservice conditions when communication distance betweentransmitter and receiver is less than 1198890 and the free spacemodule will be adopted Otherwise if the communicationdistance is beyond 1198890 two-ray ground module will be usedin which the energy that is consumed in transmitter sendingdata is in proportion to the biquadrate of the communicationdistance Therefore the trait of the module mentioned aboveis that the transmitter automatically uses different wirelesschannel module to work out the energy amount required insending data in terms of communication distance

Energy efficiency is the pivotal issue of WSNs whichrequires free space module to be used at best in the commu-nication between the transmitter and receiver for which thecommunication distance between nodes should keep withinthe distance1198890 In a clustering structure network the distancebetween cluster head usually is longer than that betweencluster head and its correspondingmember node which needcommunication radius to be less than the distance 11988902 inintracluster if the distance that is less than 1198890 is anticipant inintercluster By limiting communication distance the energyis saved at last

According to the wireless channel module defined abovethe energy module below is available

33 Energy Module The energy consumption that the trans-mitter sends 119896 bits data to the receiver with the distance 119889 is

119864119905 (119896 119889) =

119896119864elec + 1198961205761198911199041198892

119889 lt 1198890

119896119864elec + 1198961205761198981199011198894

119889 ge 1198890

(13)

The node received 119896 bits data which consumes energy asfollows

119864119903 (119896) = 119896119864elec (14)

If a node spends 119864fusion energy to aggregate one bit thenthe energy used in aggregating 119898 data packages to a singlepackage is

119864119891 (119898 119896) = 119898119896119864fusion (15)

4 CFWA Algorithm Description

Thefirst goal of ourwork is to tackle the problemof the clusterhead maldistribution which will result in unbalanced load inthe whole network and premature death of some nodes Theresolvent is to use the fuzzy weight attribute degree algorithmto establish a cluster-based routing in network layer

CFWA algorithm is composed of two phases initial-ization and operation There are several time slices in ini-tialization phase used to receive the signal from the sinknode and implement flooding to obtain the grads levelOperation phase contains setup phase and steady-state phaseIn the setup phase there are 4 subperiods including nodebroadcasting cluster head broadcasting member joiningcluster and TDMA schedule broadcasting The steady-statephase contains a few rounds and these rounds consist ofseveral frames The time structure is shown in Figure 2

41 Initialization The sink node broadcasts a beacon at acertain power the sensor node who received the beaconsignal should limit within a region at the radius of (119889max minus11988908)

After a period of delay the sink node broadcasts anotherbeacon at the maximum power of the sensor node by whichthe radio wave covers a circle region at the radius of 119889maxThe node that has received this beacon evaluates the distancebetween the sink node and itself based on the RSSI as well asgaining the grads level 0

After evaluating distance the node who received eitherbeacon turns off transceiver and goes into dormancy Thenode who only received the second beacon wakes up andstarts broadcasting its own grads level at the radius of 11988908 ata random time in the certain interval during which all nodesthat hold the same grads level will complete broadcastingtheir own grads levels and then goes to sleep again And thenodes which have never received any signals before receivethis message and set its grads level as 1 (received messageplus 1) from which the distance between the sink nodeand itself is considered as (119889max + 11988908) When a node hasreceived anymessage about grads level it goes into dormancyimmediately After the broadcasting that is implemented bythe nodes whose grads level is 0 and which has only receivedthe second beacon is ended the receivers broadcast theirgrads levels at the same radius of11988908 at a random time beforegoing into dormancy The node who receives this messagesets their grads level to 2 and goes to sleep until the timeoutof the nodes who broadcast the message ldquo1rdquo The rest maybe deduced by analogy until each node in the network hasa grads level as shown in Figure 3


Initialization

Receive 1stbroadcasting

from sink

Broadcastgrads levelnumber 1


Receive 2ndbroadcasting

from sinkNodes

broadcastingCluster-head

advertisementCluster-headsend TDMA

scheduleInform

cluster-headSlot forSlot for Slot for

Initialization

Setup Steady-state Setup Steady-stateSetup Steady-state

Round RoundRound

Setup Steady-state

Frame

Initialization

T1 T2 Tb T998400

middot middot middotmiddot middot middot

middot middot middot

node inode inode i

TR1 TR2

Figure 2 CFWA algorithm time structure diagram

TransmitterReceiver

Sink nodeSleep node

dmax

dmax minus d08

Figure 3 Flooding in initialization phase

Algorithm 1 is the pseudocode of the initialization

42 Clustering Each node broadcasts a message ⟨ID 119864119903⟩ ata certain power in a period of time 119879119887 which covers a regionat the radius of 11988902 Each node receives the messages fromneighbors and stores the information into memory after theend of broadcasting based onwhich each node calculates outneighbor number Deg the average residual energy and thecost in intracluster communication RSSIave Thus the nodecan obtain all attributes it wants

A calculation will be implemented in terms of formula(8) or (9) or (10) to obtain the absolute attribute degree 119865by each node The node whose absolute attribute degree 119865 isbigger has higher probability of being the cluster head thanthe smaller one because the former has great advantage overthe latter in the energy efficiency Then the absolute attributedegree 119865 is mapped onto the time axis before the cluster headbroadcasts by means of the timer triggering from which the

node whose absolute attribute degree is bigger broadcastscluster head information earlier

A timer 119879119894 whose time span is determined by theabsolute attribute degree 119865119894 is set for each node For thereason that the node whose absolute attribute degree isbigger broadcasts cluster head information at earlier time thefollowing equation is available

11987910158401015840

119894= (1 minus 119865119894) times 119879

1015840 (16)

Here 1198791015840 is the total time in which all cluster heads

broadcast information However data packages collision isinevitable if the nodes that hold the same absolute attributedegree value in the network implement the simultaneouscluster head broadcasting To avoid this a random numberbetween 0 and 1 is introduced to generate disturbance Aconstant 120582 is set to be 09 by which the relationship between119879119894 and the absolute attribute degree 119865119894 would not be affectedThus the improved equation is described as follows

119879119894 = 1 minus [120582 times 119865119894 + (1 minus 120582) times rand (0 1)] times 1198791015840 (17)

Equation (17) makes the absolute attribute degree valuemap onto the time axis based on which the node whoseabsolute attribute degree 119865 is bigger will have a timeoutearlier When time is up a ⟨ID⟩ package is broadcasted at theradius of 1198890 by the cluster head The nodes who can receiveand parse the package correctly will lost the chance of beingthe cluster heads if the sender is lying in the neighbor listthat is stored inmemory which is the radio coveragemethodwhich makes the cluster head distribute evenly in the wholenetwork

The cluster head broadcasting phase is finishedwhen time1198791015840 is up

43 The Establishment of Routing in Intercluster and in Intr-acluster After receiving the broadcasting of cluster headsthe member nodes select the nearest cluster head based onthe RSSI and send the join information to it The distancebetween the member node and the corresponding clusterhead is evaluated as well


void time snychronization( )int grads = 0if (receive(hello1)) flag = 1else if (receive(hello2)) lowastestimate distance between sink and itself 1198890 is the

demarcation point between free space model andtwo ray ground model RSSI(SK) and RSSI(ID)represent the received signal strength at sink nodeand this node respectivelylowastdouble dtosk = pow(100 (RSSI(SK) minus RSSI(ID) + 119909119894(sigma))(10 lowast lambda))1198890if (flag = 1) broadcast(grads)

lowast119883 and 119884 are the bandaries of deployment arealowastwhile (119889max + grads lowast11988908 lt 119889max + 119884)

If (receive(grads))

grads++119889max is the farthest distance to receive the sinkrsquos signaldtosk = 119889max + grads lowast11988908broadcast(grads)

Algorithm 1 The pseudocode implementation of the initialization

If the distance between the member node and the clusterhead is more than its distance to the sink node the membernode will communicate with the sink node directly at a fixedtime slice regardless of cluster head while going to dormancyat the rest time to save energy

Algorithm 2 shows the pseudocode of the setup phase Asfor a cluster head the nearest cluster head will be selectedto join into based on the RSSI if the distance between thesink node and the selected cluster head is shorter than thedistance between the source cluster heads and the sink nodeThe distance between the relational cluster heads is evaluatedin the same way If the distance between cluster heads is morethan the distance between the cluster head and the sink nodeor there is no cluster head nearer the sink node than itselfthis cluster head communicates directly with the sink node

The cluster head assigns a time slot for eachmember afterreceiving all join information by which a TDMA schedule isschemed The cluster heads who communicate directly withthe sink node promulgate the schedule firstly and the othercluster heads who cannot communicate directly with thesink node promulgate the schedule only when it received theschedule from its cluster head of upper level

As one of the members the cluster head communicatewith its cluster head of upper level at the appointed time slotwhen the routing in intercluster is established

The routing is simpler in intracluster The member ofnodes who go to dormancy at the rest time to save energycommunicates directly with the cluster head at the appointedtime slot

Similar to LEACH the usage of a TDMACDMA MACwill reduce inter- and intracluster collisions in CFWA familyalgorithms

44 Data Transmission The interval used in data transmis-sion is much longer than the time of setup phase so that theenergy dissipation can be reduced further Compared withthe LEACH-like algorithms CFWA family algorithms havelonger time in data transmission

At data transmission phase the member nodes sendinformation to the cluster head according to the scheduleand then go into dormancy while the cluster head must keepunder working state to receive the information coming fromits members and send the aggregated data to the next hop atthe time slot that is assigned by the cluster head of upper levelThe cluster head who communicates directly with the sinknode implements data fusion after a frame and then sendsthe aggregated data to the sink node

5 Simulation and Analysis

51 The Selection of Simulation Platform NS2 is adoptedas the simulation platform in this paper As a discreteevent simulator NS2 in which the object-oriented designtechnique is introduced and plenty of function modulesare furnished can simulate and analyze various networkprotocols and draw very intuitionistic conclusions about theperformance analysis of the system

LEACH-like algorithms such as LEACH AOW-LEACHand DCHS are simulated and compared with CFWA familyalgorithms in the same scene as the parameters are set inTable 2

52 Description Comparison From Figure 4 it is clear thatCFWA achieved more well-proportioned cluster descriptionamong the algorithms Due to not many limitations on the


broadcast(ID 119864119903)calculate(Deg) Obtain the neighbor numbercalculate(RSSIave) Obtain the average RSSI of my neighborsnormalization procedure of Deg119865(Deg) = (6479 lowast Degand2 minus 2266Deg + 3001)(119899 lowast 119864max lowast (119909

and2 minus 1017119909 + 2434))

if (CFWA 1)119865 = 1199081 lowast (119864119903119864max) + 1199082 lowast (RSSIaveTSSI) + 1199083 lowast 119865(Deg) elseif (CFWA 2)119865 = pow(119864119903119864max 1199081) + pow(RSSIaveTSSI 1199082) + pow(119865(Deg) 1199083) elseif (CFWA 3)119865 = pow(119864119903119864max 1199081) lowast pow(RSSIaveTSSI 1199082) lowast pow(119865(Deg) 1199083) 119879119894 = (09 lowast (1 minus 119865) + 01 lowast rand(0 1)) lowast 1198791015840if (receive(c1uster head) ampamp 119879119894 == now)now is the current time not receiveany advertisement

broadcast(cluster head) declare itself as cluster headheadFlag = 1 cluster head mark is set

if (1198791015840 == now)if has received some messagelowast119901 = receivedCHList[ ] load in the listif (length(receivedCHList) = 0)

currentCH = selectCH(lowast119901) select cluster headjoin(currentCH) send join information

else

sleep( )

if (HeadFlag == 1)cluster head

creatTDMA( ) generate TDMA scheduleif (currentCH == Sink) broadcast(TDMA)if (receive(TDMA)) broadcast(TDMA)

elsereceive(TDMA)

Algorithm 2 The pseudocode implementation of the setup phase

Table 2 Parameters in simulation

Parameter ValueInitial energy (J) 2Data packet size (Bytes) 500Threshold distance (119889

0) (m) 86

Packet header size (Bytes) 25119864elec (nJbit) 50119864fusion (nJbitsignal) 5120576fs (pJbitm

2) 10120576mp (pJbitm

4) 00013

radius of the clusters and cluster heads selection so thecluster distribution is casual in LEACH DCHS only limitedthe cluster heads selection on energy hence the clusterdescription is not even AOW-LEACH took the cluster heads

selection into account in which some parameters play rolesfor the even distribution of cluster However the scale of thecluster was not restricted so that the cluster description wasnot very homogeneous CFWA adopted multiple approachessuch as limiting cluster radius and selecting cluster headsaccording to several parameters to ensure the uniformclusterdescription

CFWA family algorithms are composed of CFWA 1CFWA 2 and CFWA 3 which is developed in terms of thedifferent absolute attribute degree value 119865 from (8) (9) and(10) respectively The simulation firstly takes place in thenetwork with 100 nodes in which the deployment area is100m times 100m and the sink node is located at (50 175) Thesimulation results are described in Figures 5 6 and 7

53 Performance Analysis Figure 5 denotes the relationbetween nodes number alive and runtime from whichit is obvious that CFWA family algorithms enhance 30


0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA

Figure 4 Cluster description comparison in a certain round

LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH

Figure 5 Alive nodes comparison diagram between CFWA family algorithms and LEACH-like algorithms in 100 nodes


Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH

Figure 6 Energy analysis chart between CFWA family algorithmsand LEACH-like algorithms in 100 nodes

Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH

Figure 7 Traffic comparison diagram between CFWA familyalgorithms and LEACH-like algorithms in 100 nodes

approximately more than LEACH on the total runtime of theentire network as well as 5ndash10 more than AOW LEACHand DCHSThe time of the first dead node is at 480th secondinCFWA family algorithmswhile 400th second is available inLEACH-like algorithms which is a great improvement Themain reason for this result is that the power that can coverthe circle region at the radius of 11988902 is used to broadcastinformation to neighbors in addition to themultihop routingin intercluster and the longer interval of data transmissionis adopted These measures reduce and balance the energy

Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH

Figure 8 Alive nodes comparison diagram between CFWA familyalgorithms and LEACH-like algorithms in 200 nodes

consumption of the whole network From the point of viewof the individual the energy of each node is saved and usedefficiently so that the lifetime of the network is extended

Energy analysis demonstrates that CFWA family algo-rithms have consumed similar energy at each round duringthe network operation while they have the different energydissipation at each round in LEACH-like algorithms espe-cially after the first node death as described in Figure 6The reason for this phenomenon is that the radio coveragemethod is carried out to ensure the uniform distributionof the clusters which is conducive to balancing the energydepletion of the entire network Furthermore CFWA familyalgorithms consume less energy than LEACH-like algorithmsat each round This is because the multihop routing is usedto forward data to the sink which makes cluster heads avoidsending the data to the sink directly

The traffic that is received by the sink node is shownin Figure 7 from which it is indicated that the traffic ofCFWA family algorithms is much less than that of LEACH-like algorithms The reason for the great differences of trafficin the operation of two kinds of algorithms is that clusterhead node only implements data aggregation once beforedata is sent to the sink node in LEACH-like algorithms whilemultiple data aggregations are run during the process of databeing transmitted to the sink node

The performance of the network with 200 nodes is alsoevaluated through the simulations The same parameters asthat in 100 nodes scene are used to create the simulationmodel and the results are demonstrated in Figures 8 9and 10 respectively From the charts we can clearly seethat the phenomenons emerged from the simulations in thescene with 200 nodes which is more obvious than that inthe scene of 100 nodes This is due to the increased clusternumber When nodes quantity increases the cluster number


Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


is also increased In this case multiple cluster heads transmitdata to the sink node at random time in the interval of around which will result in the severe collisions bakeoffs andretransmission of data if there is the lack of timemanagementWhen the total number of nodes in the whole network is 200the actions mentioned above will consume lots of energy andshorten the lifetime of the network Oppositely the TDMAmechanism is adopted in CFWA family algorithms to avoiddata collision so that the death time of the first node and thelifespan of the whole network of CFWA family algorithms arelonger than that of LEACH-like algorithms

Similar situation happens in the network of 500 nodesin which the deployment area is 200m times 200m and the sink

Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


node is located at (100 275) as described in Figures 11 12 and13

The performance of CFWA 1 CFWA 2 and CFWA 3 issimilar in the three scenes which denotes that the reliabilitythe stability and the scalability of CFWA family algorithmsare especially excellent

54 Parameters

541 Node Broadcasting Time 119879119887 Each node broadcastsinformation to the neighbors at the radius of 11988902 at thebeginning of each round the time span of which is the pivotal


Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


factor that may influence the usage of the energy If 119879119887 is toolarge each node will increase waiting time so as to consumeunwanted energy of idle state However the parameter isconnected with the network size If the network size is toolarge 119879119887 must be enlarged in order to avoid the collision thathappened on account of broadcasting in the limited time

542 Cluster Head Broadcasting Time 1198791015840 Cluster head

broadcasting is transmitted by radio in turn based on thetime order that is mapped by the absolute attribute degreeof its own 1198791015840 which is the total time span in the process ofthe broadcasting of cluster heads is also a significant factorthat influences energy efficiency If 1198791015840 is too small the radiocoverage method will not be implemented That is there isdelay during the propagation of radio wave If a node hasbroadcasted the cluster head advertisement the informationneeds a short period of time to transfer Just in this period acertain neighbor nodemay declare itself to be the cluster headbecause the timer has been triggered and no information isreceived In this case several cluster heads maybe lie in theadjacent regions or the same cluster The large 1198791015840 will resultin energy consumption in thewaiting time of idle state whichmakes the lifetime of the whole network shorten

55 Complexity Analysis It is assumed that there are 119899 nodesin the network and the nodes broadcast 119899⟨ID 119864119903⟩ messagesduring the cluster head selection followed by 119896 cluster headbroadcasting if 119896 cluster heads are selected all over thenetwork Even if only one cluster head can communicatedirectly with the sink node 119899 minus 1 join messages will bebroadcasted by all nodes Furthermore 119896 cluster heads willbroadcast at most 119896 TDMA schedule subsequently Thus thetotal message spending in the phase of cluster forming is119899 + 119896 + 119899 minus 1 + 119896 = 2119899 in the whole network which denotesthat the message complexity of CFWA family algorithms inthe setup phase is 119874(119899)

All nodes finish broadcastingwithin119879119887 while the timer ofeach node will stop when cluster head broadcasting interval1198791015840 is over Likewise the process of nodesrsquo joining clusters

and cluster head broadcasting TDMA schedule is also accom-plished in fixed intervalTherefore the time complexity of thealgorithm CFWA in setup phase is 119874(1)

56 Network Scalability Analysis The direct communicationwith the sink node is adopted in LEACH-like algorithmswhich will limit the network size to a great extent This ismostly because some cluster heads are far away from thesink node and cannot communicate with it even if the largestpower is used which results in the waste of the energy ofsome cluster The multihop relay is used to forward datain CFWA with the distributed algorithm that only needslocal information in cluster forming algorithmTherefore therouting which is established by CFWA algorithm is suitableto large scale wireless sensor networks

6 Conclusions

Clustering routing algorithm is an important research issuewhich will also influence the operational efficiency of net-work On the basis of analysis and comparison of some clas-sical algorithm a novel clustering routing algorithm CFWAis proposed The fuzzy weight absolute degree is introducedto make the most factors that can influence energy efficiencybecome an organic whole to determine the selection of thecluster head which is the main innovation and improvementof the classical algorithms Moreover CFWA supports datafusion both in intercluster and in intracluster which caneliminate the redundant data effectively so as to reduce thetraffic and save the energy In addition CFWA selects thenearest path to forward the aggregated data to the sink nodeat the type of multihop by comparing the distance betweennode and the corresponding cluster head and that betweennode and the sink node The simulation results show that thelifetime and energy efficiency of CFWA family algorithms isbetter than the classical algorithm

Although improvements are made in some performancethere are some limits such as time synchrony and faulttolerant in using this algorithm yet CFWA belongs to thetable-driven routing algorithm so that this protocol is mostappropriate when constantmonitoring by the sensor networkis needed

Another disadvantage of CFWA algorithm is that delaygenerating in the data transmission process from a nodeto the sink node is too long This is because data fusion isimplemented at each cluster head in the path toward the sinknode

Furthermore the ant colony optimization techniqueshould be introduced into the direct methodology in order toachieve a better cluster head distribution all over the networkand spare cluster head and path should be used to promotethe robustness in further work

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper


Acknowledgments

This work was funded in part by a grant from NationalNatural Science Foundation of China no 5307012 This workwas also partly supported by the fund of the general programof Liaoning Provincial Department of Education ScienceResearch no L2013210 and Dalian Polytechnic UniversityYouth Grants no QNJJ201307

References

[1] Z Zinonos C Chrysostomou and V Vassiliou ldquoWireless sen-sor networks mobility management using fuzzy logicrdquo Ad HocNetworks vol 16 pp 70ndash87 2014

[2] K W Sha J Gehlot and R Greve ldquoMultipath routing tech-niques in wireless sensor networks a surveyrdquoWireless PersonalCommunications vol 70 no 2 pp 807ndash829 2013

[3] B A Said E Abdellah and M Ahmed ldquoGateway and clusterhead election using fuzzy logic in heterogeneous wireless sensornetworksrdquo in Proceedings of the International Conference onMultimedia Computing and Systems (ICMCS rsquo12) pp 761ndash766Tangier Morocco May 2012

[4] L Barolli QWang E Kulla B Kamo F Xhafa andM YounasldquoA fuzzy-based simulation system for cluster-head selectionand sensor speed control in wireless sensor networksrdquo in Pro-ceedings of the 3rd International Conference on Emerging Intel-ligent Data and Web Technologies (EIDWT rsquo12) pp 16ndash22Bucharest Romania September 2012

[5] S Maurya and A K Daniel ldquoHybrid routing approach forheterogeneous wireless sensor networks using fuzzy logic tech-niquerdquo in Proceedings of the 4th International Conference onAdvanced Computing and Communication Technologies (ACCTrsquo14) pp 202ndash207 IEEE Rohtak India February 2014

[6] P Kumari M P Singh and P Kumar ldquoSurvey of clusteringalgorithms using fuzzy logic in wireless sensor networkrdquo in Pro-ceedings of the International Conference on Energy Efficient Tech-nologies for Sustainability (ICEETS rsquo13) pp 924ndash928 April 2013

[7] J N Al-Karaki and A E Kamal ldquoA taxonomy of routingtechniques in wireless sensor networksrdquo inHandbook of SensorNetworks CompactWireless andWired Sensing SystemsM Ilyasand I Mahgoub Eds pp 116ndash139 CRC Press 2005

[8] W R Heinzelman A P Chandrakasan and H BalakrishnanldquoEnergy-efficient communication protocol for wireless micro-sensor networksrdquo in Proceedings of the 33rd Annual HawaiiInternational Conference on System Siences (HICSS rsquo00) pp3005ndash3014 January 2000

[9] A Manjeshwar and D P Agarwal ldquoTEEN a routing protocolfor enhanced efficiency inwireless sensor networksrdquo inProceed-ings of the 15th International Parallel and Distributed ProcessingSymposium pp 2009ndash2015 IEEE San Francisco Calif USAApril 2000

[10] A Manjeshwar and D P Agarwal ldquoAPTEEN a hybrid protocolfor efficient routing and comprehensive information retrievalin wireless sensor networksrdquo in Proceedings of the 2nd Interna-tionalWorkshop on Parallel and Distributed Computing Issues inWireless Networks and Mobile Computing (IPDPS rsquo02) pp 195ndash202 Lauderdale Fla USA April 2002

[11] M J Handy M Haase and D Timmermann ldquoLow energyadaptive clustering hierarchy with deterministic cluster-headselectionrdquo in Proceedings of the 4th IEEE Conference on Mobileand Wireless Communications Networks pp 368ndash372 Stock-holm Sweden 2002

[12] A K Parekh ldquoSelecting routers in ad-hoc wireless networksrdquoin Proceedings of the the SBTIEEE International Telecommuni-cations Symposium (ITS rsquo94) pp 420ndash424 Rio de Janeiro BrazilAugust 1994

[13] C R Lin andM Gerla ldquoAdaptive clustering for mobile wirelessnetworksrdquo IEEE Journal on Selected Areas in Communicationsvol 15 no 7 pp 1265ndash1275 1997

[14] O Younis and S Fahmy ldquoHeed a hybrid energy-efficientdistributed clustering approach for ad hoc sensor networksrdquoIEEE Transactions on Mobile Computing vol 3 no 4 pp 366ndash379 2004

[15] M Chatterjee S K Das and D Turgut ldquoWCA a weightedclustering algorithm for mobile ad hoc networksrdquo Journal ofCluster Computing vol 5 no 2 pp 193ndash204 2002

[16] J-W Zhang Y-Y Ji J-J Zhang and C-L Yu ldquoA weightedclustering algorithm based routing protocol in wireless sensornetworksrdquo in Proceedings of the ISECS International Colloquiumon Computing Communication Control and Management(CCCM rsquo08) pp 599ndash602 IEEE Guangzhou China August2008

[17] H-Q Huang D-Y Yao J Shen K Ma and H-T Liu ldquoMulti-weight based clustering algorithm forwireless sensor networksrdquoJournal of Electronics amp Information Technology vol 30 no 6pp 1489ndash1492 2008

[18] Y Y Ji J W Zhang and C L Yu ldquoAn improvement routingprotocol by weighted clustering algorithm in WSNrdquo Journal ofHangzhou Dianzi University vol 28 no 6 pp 29ndash32 2008

[19] B Cai and X D Chen ldquoClustering algorithm based on auto-matic on-demand weighted for sensor networksrdquo Microelec-tronics amp Computer vol 25 no 11 pp 129ndash132 2008

[20] X C Huang A study on theories and methodologies for fuzzymulti-objective decision makings with their applications [PhDdissertation] Dalian University of Technology Dalian China2003

[21] S Y Chen Engineering Fussy Theories and Application Press ofDalian University of Technology Dalian China 1998

[22] W B Heinzelman A P Chandrakasan and H Balakrish-nan ldquoAn application-specific protocol architecture for wirelessmicrosensor networksrdquo IEEE Transactions onWireless Commu-nications vol 1 no 4 pp 660ndash670 2002

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



scalability lifetime and energy efficiency [7] Thus cluster-based routing takes great advantage over the plat-based oneat above performances However the disadvantage of cluster-based routing is that the cluster head is so vital that it becomesthe bottleneck of the entire network Therefore the selectionof the cluster head will influence the performance of theentire network The existing clustering algorithms differ onthe criterium for the selection of the cluster heads Accordingto the current research findings the selection method of thecluster head can be divided into several categories below

The classical routing protocols that are based on k-meansclustering such as LEACH [8] TEEN [9] and APTEEN [10]select cluster head based on a random acquired value Ifthis value is less than a certain threshold the nodes willbe the cluster head Whereas because of the randomicityduring the selection the selected cluster head is prone tobe distributed improperly and unevenly this could cause theuneven distribution of the traffic flow in different cluster headnodes One of the direct consequences is that some clusterheads exhaust energy at the mean time the performance ofthe entire network is affected

Some distributed routing protocols based on a certainattribute are proposed in DCHS [11] HCDA (the Highest-ConnectivityDegreeAlgorithm) [12] andACMWN[13]Theattributes that can determine cluster head selection includeresidual energy neighbors number the cost that communi-cate in intracluster and the distance between the node andthe sink node and ID Because only one attribute is taken intoaccount in these protocols the selected cluster head cannotbe the most suitable node Although the rationality of thecluster head selection is improved to a certain extent certainproblems such as the unevenly distributed cluster head andthe imbalance load remain unsolved

Multiattribute cluster head selection protocol such asHEED [14] and WCA [15] which use several attribute todetermine the cluster head are greatly favored due to the con-sideration of various factors The advantage of multiattributecluster head selection is that a better partition of clustercan be obtained The two protocols both adopt successivescreeningmethod to determine the cluster head bywhich thefinite iteration must be implemented The major drawbackof the former is that distributed algorithm makes each nodeunaware of global information so that some nodes may notjoin any clusters while the latter need to iterate many times ifmany attributes are used to gain a better performance whichwill increase time complexity and consume more energy

WCA-LEACH [16] MWBC [17] WCA-GSEN [18] andAOW-LEACH [19] combine multifactors such as residualenergy communication cost and neighbor nodes number inorder to avoid the randomicity in the cluster head selectionof LEACH However all the algorithms above determineweight of each factor using trial and error method which willinfluence the performance of the whole protocol

From the analysis and comparison mentioned abovemultiattribute cluster head selection can obtain the unparal-leled rationality in partition of cluster therefore we considerthe residual energy neighbors number the cost that commu-nicate in intracluster and the distance between the node andthe sink node as the attribute to propose a new clustering

routing algorithm using fuzzy weight multiattribute (CFWA)to determine cluster head selection by which the energycan be saved and the lifetime of the whole network will beextended

2 Conform the Weight of Attribute

To save energy and balance load the residual energy is themost crucial factor of the attributes during the process ofcluster head selection The cost that is used in the com-munication in intracluster and the neighbors number alsoinfluence cluster head selection Nevertheless the distancebetween the node and the sink node will not be considereddue to the fact that uniform distribution of the cluster head isrequired by energy-efficient cluster-based algorithm In thispaper the direct method [20] based on the abutting objectrelative membership degree in engineering fuzzy theory andintelligence decision-making is adopted in order to confirmthe proportion of each attribute that is hold during clusterhead selection

Definition 1 Compare the member119874119896 with another member119874119897 on duality about weightiness in the object set119874 When119874119896is more important than 119874119897

05 lt 120573119896119897 ⩽ 1 (1)

when 119874119897 is more important than 119874119896

0 ⩽ 120573119896119897 lt 05

120573119896119897 = 1 minus 120573119897119896

(2)

when 119874119896 has the same importance as 119874119897

120573119896119897 = 05

especially 120573119896119896 = 05

(3)

where120573119896119897 is named relatively weightiness fuzzy value betweenthe object 119874119897 and 119874119896 Particularly if the object sequencingabout weightiness is 1198741 ≺ 1198742 ≺ sdot sdot sdot ≺ 119874119898 12057311989611198961+1 (1198961 =1 2 119898 minus 1) is defined as the abutting object relativelyweightiness fuzzy value

Assumption 2 In the available attributes it is assumed thatresidual energy (119864119903) is more significant than the cost thatcommunicate in intracluster (Cost) and the latter is moreimportant than neighbors number (Deg)

That is residual energy has the unexampled importancethan the cost that communicate in intracluster while the latteris more important than neighbors number ratherish Therelevant fuzzy value that 120573119864119903Cost is 1 and 120573CostDeg is 055 canbe found out based on Table 1 [21]

Based on the assumption about relative significance fuzzyscale value provided that the object 119864119903 is more importantthan Cost 120573119864119903Cost is the corresponding significance degreewhen just comparing objects 119864119903 and Cost of which bench-mark is 119864119903 the more important one between these twoobjects Because 120573119864119903119864119903 which is the fuzzy scale value that






120573119864119903

120573Cost=

1

15 minus 120573119864119903Cost (4)


1205961015840

119864119903= 1

1205961015840

Cost = 1205961015840

119864119903(15 minus 1) = 05

1205961015840

Deg = 1205961015840

Cost (15 minus 055) = 0475

(5)


120596 = (120596119864119903 120596Cost 120596Deg) = (05063 02532 02405) (6)





20

25

30

35

40

45

50

55

60

65

Aver

age e

nerg

y di

ssip

atio

n pe

r rou

nd (J

)




119865 (Deg) = 6479 times 1199092minus 2266119909 + 3001


(7)



1198651 = 120596119864119903 sdot (119864119903

119864max) + 120596Cost sdot (

RSSIaveTSSI

) + 120596Deg

sdot 119865 (Deg) (8)


1198652 = (119864119903

119864max)

120596119864119903

+ (RSSIaveTSSI

)

120596Cost

+ 119865 (Deg)120596Deg (9)


1198653 = (119864119903

119864max)

120596119864119903

sdot (RSSIaveTSSI

)

120596Cost

sdot 119865 (Deg)120596Deg (10)



3 System Module

31 Network Module







RSSI = 119860 minus 10119899 log 10 (119889) (12)




whole network





119864119905 (119896 119889) =

119896119864elec + 1198961205761198911199041198892

119889 lt 1198890

119896119864elec + 1198961205761198981199011198894

119889 ge 1198890

(13)


119864119903 (119896) = 119896119864elec (14)


119864119891 (119898 119896) = 119898119896119864fusion (15)








Initialization


from sink




from sinkNodes



scheduleInform


Initialization


Round RoundRound

Setup Steady-state

Frame

Initialization

T1 T2 Tb T998400



node inode inode i

TR1 TR2


TransmitterReceiver

Sink nodeSleep node

dmax

dmax minus d08







11987910158401015840

119894= (1 minus 119865119894) times 119879

1015840 (16)











If (receive(grads))

















and2 minus 1017119909 + 2434))





else

sleep( )



elsereceive(TDMA)




0) (m) 86


2) 10120576mp (pJbitm

4) 00013






0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA


LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH



Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation














120573119864119903

120573Cost=

1

15 minus 120573119864119903Cost (4)


1205961015840

119864119903= 1

1205961015840

Cost = 1205961015840

119864119903(15 minus 1) = 05

1205961015840

Deg = 1205961015840

Cost (15 minus 055) = 0475

(5)


120596 = (120596119864119903 120596Cost 120596Deg) = (05063 02532 02405) (6)





20

25

30

35

40

45

50

55

60

65

Aver

age e

nerg

y di

ssip

atio

n pe

r rou

nd (J

)




119865 (Deg) = 6479 times 1199092minus 2266119909 + 3001


(7)



1198651 = 120596119864119903 sdot (119864119903

119864max) + 120596Cost sdot (

RSSIaveTSSI

) + 120596Deg

sdot 119865 (Deg) (8)


1198652 = (119864119903

119864max)

120596119864119903

+ (RSSIaveTSSI

)

120596Cost

+ 119865 (Deg)120596Deg (9)


1198653 = (119864119903

119864max)

120596119864119903

sdot (RSSIaveTSSI

)

120596Cost

sdot 119865 (Deg)120596Deg (10)



3 System Module

31 Network Module







RSSI = 119860 minus 10119899 log 10 (119889) (12)




whole network





119864119905 (119896 119889) =

119896119864elec + 1198961205761198911199041198892

119889 lt 1198890

119896119864elec + 1198961205761198981199011198894

119889 ge 1198890

(13)


119864119903 (119896) = 119896119864elec (14)


119864119891 (119898 119896) = 119898119896119864fusion (15)








Initialization


from sink




from sinkNodes



scheduleInform


Initialization


Round RoundRound

Setup Steady-state

Frame

Initialization

T1 T2 Tb T998400



node inode inode i

TR1 TR2


TransmitterReceiver

Sink nodeSleep node

dmax

dmax minus d08







11987910158401015840

119894= (1 minus 119865119894) times 119879

1015840 (16)











If (receive(grads))

















and2 minus 1017119909 + 2434))





else

sleep( )



elsereceive(TDMA)




0) (m) 86


2) 10120576mp (pJbitm

4) 00013






0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA


LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH



Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation














RSSI = 119860 minus 10119899 log 10 (119889) (12)




whole network





119864119905 (119896 119889) =

119896119864elec + 1198961205761198911199041198892

119889 lt 1198890

119896119864elec + 1198961205761198981199011198894

119889 ge 1198890

(13)


119864119903 (119896) = 119896119864elec (14)


119864119891 (119898 119896) = 119898119896119864fusion (15)








Initialization


from sink




from sinkNodes



scheduleInform


Initialization


Round RoundRound

Setup Steady-state

Frame

Initialization

T1 T2 Tb T998400



node inode inode i

TR1 TR2


TransmitterReceiver

Sink nodeSleep node

dmax

dmax minus d08







11987910158401015840

119894= (1 minus 119865119894) times 119879

1015840 (16)











If (receive(grads))

















and2 minus 1017119909 + 2434))





else

sleep( )



elsereceive(TDMA)




0) (m) 86


2) 10120576mp (pJbitm

4) 00013






0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA


LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH



Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Initialization


from sink




from sinkNodes



scheduleInform


Initialization


Round RoundRound

Setup Steady-state

Frame

Initialization

T1 T2 Tb T998400



node inode inode i

TR1 TR2


TransmitterReceiver

Sink nodeSleep node

dmax

dmax minus d08







11987910158401015840

119894= (1 minus 119865119894) times 119879

1015840 (16)











If (receive(grads))

















and2 minus 1017119909 + 2434))





else

sleep( )



elsereceive(TDMA)




0) (m) 86


2) 10120576mp (pJbitm

4) 00013






0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA


LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH



Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













If (receive(grads))

















and2 minus 1017119909 + 2434))





else

sleep( )



elsereceive(TDMA)




0) (m) 86


2) 10120576mp (pJbitm

4) 00013






0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA


LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH



Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











and2 minus 1017119909 + 2434))





else

sleep( )



elsereceive(TDMA)




0) (m) 86


2) 10120576mp (pJbitm

4) 00013






0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA


LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH



Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(a) LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(b) DCHS

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(c) AOW LEACH

0102030405060708090

100

10 20 30 40 50 60 70 80 90 1000(d) CFWA


LEACH

DCHS

Alive node analysis

CFWA_1CFWA_2CFWA_3

0

10

20

30

40

50

60

70

80

90

100

Aliv

e nod

e num

ber

100 200 300 400 500 600 7000Time (s)

AOW_LEACH



Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Energy analysis

0

20

40

60

80

100

120

140

160

180

200

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

Traffi

c

100 200 300 400 500 600 7000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH



Alive node analysis

0

20

40

60

80

100

120

140

160

180

200

Aliv

e nod

e num

ber

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH







Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Energy analysis

0

40

80

120

160

200

240

280

320

360

400

Ener

gy co

nsum

ptio

n (J

)

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Data analysis

0250050007500

1000012500150001750020000225002500027500300003250035000

Traffi

c

100 200 300 400 500 600 700 800 9000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




Alive node analysis

0

50

100

150

200

250

300

350

400

450

500

Aliv

e nod

e num

ber

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH


Alive node analysis

0

100

200

300

400

500

600

700

800

900

1000

Ener

gy co

nsum

ptio

n (J

)

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH




54 Parameters



Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Data analysis

0

2500

5000

7500

10000

12500

15000

17500

20000

Traffi

c

200 400 600 800 1000 12000Time (s)

LEACH

DCHS

CFWA_1CFWA_2CFWA_3

AOW_LEACH









6 Conclusions








Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Acknowledgments


References

























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









research article fuzzy weight cluster-based routing...

Documents