06782312.pdf

732 IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 61, NO. 2, APRIL 2014

A General Self-Organized Tree-BasedEnergy-Balance Routing Protocolfor Wireless Sensor Network

Zhao Han, Jie Wu, Member, IEEE, Jie Zhang, Liefeng Liu, and Kaiyun Tian

AbstractWireless sensor network (WSN) is a system composedof a large number of low-cost micro-sensors. This network is usedto collect and send various kinds of messages to a base station(BS). WSN consists of low-cost nodes with limited battery power,and the battery replacement is not easy for WSN with thousandsof physically embedded nodes, which means energy efficientrouting protocol should be employed to offer a long-life work time.To achieve the aim, we need not only to minimize total energyconsumption but also to balance WSN load. Researchers haveproposed many protocols such as LEACH, HEED, PEGASIS,TBC and PEDAP. In this paper, we propose a General Self-Or-ganized Tree-Based Energy-Balance routing protocol (GSTEB)which builds a routing tree using a process where, for each round,BS assigns a root node and broadcasts this selection to all sensornodes. Subsequently, each node selects its parent by consideringonly itself and its neighbors information, thus making GSTEBa dynamic protocol. Simulation results show that GSTEB hasa better performance than other protocols in balancing energyconsumption, thus prolonging the lifetime of WSN.

Index TermsEnergy-balance, network lifetime, routing pro-tocol, self-organized, wireless sensor network.

I. INTRODUCTION

W ITH the advances in Micro-Electro-Mechanical Sys-tems (MEMS)-based sensor technology, low-powerdigital electronics and low-power wireless communication[1], [2], [3], it is now possible to produce wireless sensornodes in quantity at low cost. Although these sensor nodes arenot as powerful or accurate as their expensive macro-sensorcounterparts, we are able to build a high quality, fault-tolerantsensor network by making thousands of sensor nodes worktogether. Through the cooperation of wireless sensor nodes,WSN collects large amounts of information and sends themto the Base Station (BS). WSN has a wide range of potentialapplications [10], including military surveillance, disaster pre-diction, environment monitoring, etc. Thus it has become one

Manuscript received June 30, 2012; revised February 09, 2013; acceptedFebruary 25, 2014. Date of publication April 02, 2014; date of current ver-sion April 10, 2014. This work was supported in part by Major National Sci-ence and Technology Special Program of China under Grant 2011ZX05008-005-61.The authors are with Department of Modern Physics and State Key

Laboratory of Particle Detection and Electronics, University of Sci-ence and Technology of China, Hefei 230026, Anhui, China (e-mail:[email protected]; [email protected]; [email protected];[email protected]; [email protected]).Color versions of one or more of the figures in this paper are available online

at http://ieeexplore.ieee.org.Digital Object Identifier 10.1109/TNS.2014.2309351

of the most important research fields and has aroused extensiveresearch interest.Generally, wireless sensor nodes are deployed randomly and

densely in a target region, especially where the physical envi-ronment is so harsh that the macro-sensor counterparts cannotbe deployed. After deployment, the network cannot work prop-erly unless there is sufficient battery power. In general, WSNmay produce quite a substantial amount of data, so if data fu-sion could be used, the throughput could be reduced [4]. Be-cause sensor nodes are deployed densely, WSN might generateredundant data from multiple nodes, and the redundant data canbe combined to reduce transmission. Many well-known proto-cols implement data fusion, but almost all of them assume thatthe length of the message transmitted by each relay node shouldbe constant, i.e., each node transmits the same volume of datano matter how much data it receives from its child nodes [12].PEGASIS [7], PEDAP [8] and TBC [17] are typical protocolsbased on this assumption and perform far better than LEACH[4], [5] and HEED [6] in this case. However, there are quitea few applications in which the length of the message trans-mitted by a parent node depends not only on the length of itsown, but also on the lengths of the messages received from itschild nodes. In an extreme case, the relay node should transmitthe length of the message which is the sum of its own senseddata and the received data from its children [12].Energy consumption of a node is due to either useful or

wasteful operations. The useful operations include transmit-ting or receiving data messages, and processing requests. On theother hand, the wasteful consumption is due to the operation ofconstructing routing tree, overhearing, retransmitting becauseof harsh environment, dealing with redundant broadcast over-head messages, and idle listening to the media.In this paper, we propose a General Self-Organized Tree-

based Energy Balance routing protocol (GSTEB).We consider asituation in which the network collects information periodicallyfrom a terrain where each node continually senses the environ-ment and sends the data back to BS [11]. Normally there are twodefinitions for network lifetime:a) The time from the start of the network operation to thedeath of the first node in the network [13].

b) The time from the start of the network operation to thedeath of the last node in the network.

In this paper, we adopt the first definition. Moreover, we con-sider two extreme cases in data fusion:Case (1) The data between any sensor nodes can be totally

fused. Each node transmits the same volume of

0018-9499 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

HAN et al.: GENERAL SELF-ORGANIZED TREE-BASED ENERGY-BALANCE 733

data no matter how much data it receives from itschildren.

Case (2) The data cant be fused. The length of message trans-mitted by each relay node is the sum of its ownsensed data and received data from its children.

The remainder of the paper is organized as follows: Section IIreviews related works. The network and radio models of ourproposal are discussed in Section III. Section IV describes thearchitectures and details of GSTEB. In Section V we presentour simulations in contrast to the simulations of other knownprotocols. Finally, Section VI concludes the paper.

II. RELATED WORKS

A main task of WSN is to periodically collect information ofthe interested area and transmit the information to BS. A simpleapproach to fulfilling this task is that each sensor node transmitsdata directly to BS. However, when BS is located far away fromthe target area, the sensor nodes will die quickly due to muchenergy consumption. On the other hand, since the distances be-tween each node and BS are different, direct transmission leadsto unbalanced energy consumption. To solve these problems,many protocols have been proposed. Of the protocols proposed,hierarchical protocols such as LEACH, HEED, PEGASIS, TBCand PEDAP can achieve satisfactory solutions.In LEACH [4], [5], for the entire network, nodes selected ac-

cording to a fraction p from all sensor nodes are chosen to serveas cluster heads (CHs), where p is a design parameter. The op-erations of LEACH are divided into several rounds. Each roundincludes a setup phase and a steady-state phase. During the setupphase, each node will decide whether to become a CH or not ac-cording to a predefined criterion. After CHs are chosen, each ofother nodes will select its own CH and join the cluster accordingto the power of many received broadcast messages. Each nodewill choose the nearest CH. During the steady-state phase, CHsfuse the data received from their cluster members and send thefused data to BS by single-hop communication. LEACH usesrandomization to rotate CHs for each round in order to evenlydistribute the energy consumption. So LEACH can reduce theamount of data directly transmitted to BS and balance WSNload, thus achieving a factor of 8 times improvement comparedwith direct transmission.In [6], the authors proposed a hybrid, energy-efficient, dis-

tributed clustering algorithm (HEED). HEED is an improve-ment of LEACH on the manner of CH choosing. In each round,HEED selects CHs according to the residual energy of eachnode and a secondary parameter such as nodes proximity totheir neighbors or nodes degrees. By iterations and competition,HEED ensures only one CH within a certain range, so uniformCHs distribution is achieved across the network. Compared withLEACH, HEED effectively prolongs network lifetime and issuitable for situations such as where each node has different ini-tial energy.For Case1, LEACH and HEED greatly reduce total energy

consumption. However, LEACH and HEED consume energyheavily in the head nodes, which makes the head nodes diequickly. S. Lindsey et al. proposed an algorithm related toLEACH, and it is called PEGASIS [7]. PEGASIS is a nearly

optimal power efficient protocol which uses GREEDY algo-rithm to make all the sensor nodes in the network form a chain.In PEGASIS, the (i mod N)th node is chosen to be a leader andthe leader is the only one which needs to communicate withBS in round i. N is the total amount of nodes. Data is collectedby starting from both endpoints of the chain, and transmittedalong the chain, and fused each time it transmits from one nodeto the next until it reaches the leader. So PEGASIS sharplyreduces the total amount of data for long-distance transmissionand achieves a better performance than LEACH by 100% to300% in terms of network lifetime.Tree-Based Clustering (TBC) [17] is also an improved pro-

tocol of LEACH. It forms several clusters in the same way asLEACH, and each cluster has a cluster-head (CH). The nodeswithin a cluster construct a routing tree where the cluster-headis the root of it. For tree configuration, the cluster-head uses thedistance information between themember nodes and itself. Eachnode is location-aware, it can estimate the distance between theroot and itself. Every cluster is divided into some levels. Thedistance of a node to the root is the basis for determining itslevel in the cluster. The cluster-head is at level-0(root) and anode in level will choose the node in and nearestto itself as its parent node. Data transfer simultaneously happensbetween the nodes in two neighboring levels, and each nodefuses the received data and transmits it to its parent. TBC isan excellent protocol in which each node records the informa-tion of its neighbors and builds topography through computing,which is similar to GSTEB. But some cluster-heads in the net-work consumemore energy than other nodes when BS is locatedfar away.PEDAP [8] is a tree-based routing protocol that makes all the

nodes form a minimum spanning tree, which costs minimumenergy for data transmitting. It also has another version calledPEDAP-PA which slightly increases energy for data transmit-ting but balances energy consumption per node. PEDAP has thesame network assumptions as PEGASIS and uses data fusion.However, both PEDAP and PEDAP-PA are protocols that needBS to build the topography which will cause a large amount ofenergy waste. This is because if the network needs BS to buildthe topography, BS should send a lot of information to the sensornodes, including what time is the Time Division Multiple Ac-cess (TDMA) slot, who are their child nodes and who are theirparent nodes. This kind of information exchanging will cause alot of energy to be wasted or will cause a long delay.

III. NETWORK AND RADIO MODEL

In our work, we assume that the system model has the fol-lowing properties: sensor nodes are randomly distributed in the square fieldand there is only one BS deployed far away from the area.

Sensor nodes are stationary and energy constrained. Oncedeployed, they will keep operating until their energy isexhausted.

BS is stationary, but BS is not energy constrained. All sensor nodes have power control capabilities, eachnode can change the power level and communicate withBS directly.


Sensor nodes are location-aware. A sensor node can getits location information through other mechanisms such asGPS or position algorithms.

Each node has its unique identifier (ID).In order to compare the performance of GSTEB with the per-

formance of other protocols, we use different radio models forCase1 and Case2.Since Case1 is the typical characteristic of PEGASIS [7] and

PEDAP [8], we use the same radio model as in PEGASIS anal-ysis [4], [5], [7], [8], which makes it easier to verify our simula-tion results and compare the performance of GSTEB with thatof PEGASIS. In this model, the energy dissipation of the radiocaused by running the transmitter or receiver circuitry equals

nJ/bit and the energy dissipation of the radio causedby running the transmit amplifier equals pJ/bit/m .It is also assumed that a path loss due to free-space propa-gation model is used. The energy consumption of transmittinga k-bit packet to a distance d and receiving that packet is:

TransmittingReceiving

For Case2, we use the same model as in HEED [6], whichmakes it easier to verify the simulation results and comparethe performance of GSTEB with that of HEED. This modeluses both the free-space propagation model and two-ray groundpropagation model to approximate the path loss due to wirelesschannel transmission. When , the free-space propaga-tion model is employed and uses pJ/bit/m for thetransmitter amplifier. When , the two-ray ground prop-agation model which leads to a path loss is employed anduses pJ/bit/m for the transmitter amplifier. isa threshold transmission distance which can be computed by:

square root .For both cases, the medium is assumed to be symmetric so

that the energy required for transmitting a message from nodeA to node B or from node B to node A is the same.

IV. GENERAL SELF-ORGANIZED TREE-BASEDENERGY-BALANCE ROUTING PROTOCOL

The main aim of GSTEB is to achieve a longer network life-time for different applications. In each round, BS assigns a rootnode and broadcasts its ID and its coordinates to all sensornodes. Then the network computes the path either by transmit-ting the path information from BS to sensor nodes or by havingthe same tree structure being dynamically and individually builtby each node. For both cases, GSTEB can change the root andreconstruct the routing tree with short delay and low energyconsumption. Therefore a better balanced load is achieved com-pared with the protocols mentioned in Section II.The operation of GSTEB is divided into Initial Phase, Tree

Constructing Phase, Self-Organized Data Collecting and Trans-mitting Phase, and Information Exchanging Phase.

A. Initial PhaseIn Initial Phase, the network parameters are initialized. Initial

Phase is divided into three steps.Step 1: When Initial Phase begins, BS broadcasts a packet to

all the nodes to inform them of beginning time, the

length of time slot and the number of nodes N.Whenall the nodes receive the packet, they will computetheir own energy-level (EL) using function:

EL is a parameter for load balance, and it is an esti-mated energy value rather than a true one and onlyused in Case2, i is the ID of each node, and isa constant which reflects the minimum energy unitand can be changed depending on our demands.

Step 2: Each node sends its packet in a circle with a certainradius during its own time slot after Step 1. Forexample, in the time slot, the node whose ID is iwill send out its packet. This packet contains a pre-amble and the information such as coordinates andEL of node i. All the other nodes during this time slotwill monitor the channel, and if some of them arethe neighbors of node i, they can receive this packetand record the information of node i in memory. Thenodes which are not in the range of cant mon-itor the preamble in this time slot, so they can knowthey are not the neighbors of node i and will turnoff their radios, then switch to sleep mode to saveenergy. After all nodes send their information, eachnode records a table in their memory which containsthe information of all its neighbors.

Step 3: Each node sends a packet which contains all itsneighbors information during its own time slotwhen Step 2 is over. Then its neighbors can receivethis packet and record the information in memory.The length of time slots in Steps 2 and 3 is prede-fined, thus when time is up, each node has sent itsinformation before Initial Phase ended. After Ini-tial Phase, each node records two tables in memorywhich contain the information of all its neighborsand its neighbors neighbors. These two tables aredefined as Table I and Table II. Each node works ac-cording to them in the following phases.

Initial Phase is a significant preparation for the next phases.After Initial Phase, GSTEB operates in rounds. For GSTEBand all other protocols mentioned, the round has the samemeaning. In a round, the routing tree may need to be rebuiltand each sensor node generates a DATA_PAK that needs to besent to BS. When BS receives the data of all sensor nodes, around ended. Round is not a real time measurement unit, but itreflects the ability for transmitting the collected data for sensors,so round is a suitable time measurement unit for WSN lifetime.Each round contains three phases, including Tree ConstructingPhase, Self-Organized Data Collecting and Transmitting Phase,and Information Exchanging Phase.

B. Tree Constructing Phase

Within each round, GSTEB performs the following steps tobuild a routing tree. Between Case1 and Case2 there are somedifferences in the steps of routing tree constructing:Step 1: BS assigns a node as root and broadcasts root ID and

root coordinates to all sensor nodes.


TABLE INETWORK LIFETIMES OF DIFFERENT SCHEMES

For Case1, because data fusion technique is im-plemented, only one node which communicates di-rectly with BS can transmit all the data with the samelength as its own, which results in much less energyconsumption. In order to balance the network loadfor Case1, in each round, a node with the largestresidual energy is chosen as root. The root collectsthe data of all sensors and transmits the fused datato BS over long distance.For Case2, because data cant be fused, it will notsave the energy for data transmitting by makingfewer nodes communicate directly with BS. Whenone of the sensor nodes collects all the data andsends it to BS, it would deplete its energy quickly.In this case BS always assigns itself as root.

Step 2: Each node tries to select a parent in its neighborsusing EL and coordinates which are recorded inTable I. The selection criteria are:1) For both Case1 and Case2, for a sensor node,the distance between its parent node and the rootshould be shorter than that between itself and theroot.

2) For Case1, each node chooses a neighbor thatsatisfies criterion 1 and is the nearest to itself asits parent. And if the node cant find a neighborwhich satisfies criterion 1, it selects the root asits parent.

3) For Case2, the process of Tree ConstructingPhase can be regarded as an iterative algorithm.Besides criterion 1, for a sensor node, only thenodes with the largest EL of all its neighborsand itself can act as relay nodes. If the sensornode itself has the largest EL, it can also be con-sidered to be an imaginary relay node. Choosingthe parent node from all the relay nodes is basedon energy consumptions. Any of these con-sumptions is the sum of consumption from thesensor node to a relay node and that from therelay node to BS. The relay node which causesminimum consumption will be chosen as theparent node. It is true that this relay node shouldchoose its parent node in the same way. So apath with minimum consumption is found byiterations. And by using EL, GSTEB choosesthe nodes with more residual energy to transmit

Fig. 1. Topography generated if each node chooses the nearest as parent.

data for long distance. If the sensor node cannotfind a suitable parent node, it will transmit itsdata directly to BS.

Step 3: Because every node chooses the parent from itsneighbors and every node records its neighborsneighbors information in Table II, each node canknow all its neighbors parent nodes by computing,and it can also know all its child nodes. If a nodehas no child node, it defines itself as a leaf node,from which the data transmitting begins.

As discussed above, for Case1, because each packet sent tothe parent nodes will be fused, the minimum energy consump-tion can be achieved if each node chooses the node nearest to it.But if all nodes choose their nearest neighbors, the network maynot be able to build a tree. Fig. 1 shows a network of 100 nodesin this situation. We can find that some clusters are formed, butthey cannot connect with others. Thus in GSTEB, we use crite-rion 1 in Case1 to limit the search direction. By this approach, arouting tree is constructed and some nodes still have the possi-bility of connecting to their nearest neighbors. For Case2, crite-rion 1 should also be obeyed and this criterion helps to save theenergy for data transmitting to a certain extent.To build a routing tree, for Case1, each node follows the steps.

But for Case2, we use BS to compute the topography. Eventhough we can fulfill this work without the control of BS, a largeamount of energy is wasted in the next phase.

C. Self-Organized Data Collecting and Transmitting Phase

After the routing tree is constructed, each sensor node col-lects information to generate a DATA_PKT which needs to betransmitted to BS.For Case1, TDMA and Frequency Hopping Spread Spec-

trum (FHSS) are both applied. This phase is divided into severalTDMA time slots. In a time slot, only the leaf nodes try to sendtheir DATA_PKTs. After a node receives all the data from itschild nodes, this node itself serves as a leaf node and tries tosend the fused data in the next time slot.


Fig. 2. Process of a time slot in Self-Organized Data Collecting and Transmitting Phase for Case1. stands for leaf nodes, which need to transmit data in thetime slot. stands for other nodes, they are parent nodes. All the leaf nodes try to send their DATA_PKTs in a time slot which divide all other nodes into threesituations (see the following paragraphs). This figure shows what all the nodes do for the three situations.

Each node knows the ID of its parent node. In each time slot,in order to reduce communication interference, we apply FHSSin which each child node communicates with its parent nodeusing the frequency hopping sequence determined by its parentnode ID. Each TDMA time slot is divided into three segmentsas follows (see Fig. 2).Segment1: The first segment is used to check if there is com-

munication interference for a parent node. In this segment, eachleaf node sends a beacon which contains its ID to its parent nodeat the same time.

Three situations may occur and they divide all the parentnodes into three kinds. For the first situation, if no leaf nodeneeds to transmit data to the parent node in this time slot, it re-ceives nothing. For the second situation, if more than one leafnode needs to transmit data to the parent node, it receives anincorrect beacon. For the third situation, if only one leaf nodeneeds to transmit data to the parent node, it receives a correctbeacon. The operation of the second segment depends on thethree situations.


Segment 2: During the second segment, the leaf nodes whichcan transmit their data are confirmed. For the first situation, theparent node turns to sleep mode until next time slot starts. Forthe second situation, the parent node sends a control packet toall its child nodes. This control packet chooses one of its childnodes to transmit data in the next segment. For the third situ-ation, the parent node sends a control packet to this leaf node.This control packet tells this leaf node to transmit data in thenext segment.Segment 3: The permitted leaf nodes send their data to their

parent nodes, while other leaf nodes turn to sleep mode. Theprocess in one time slot is shown in Fig. 2.For Case2, each node chooses its parent by considering not

the distance but the total energy consumption. In our simula-tion results, we will show that there may be many leaf nodessharing one parent node in one time slot. If all the leaf nodes tryto transmit their data at the same time, the data messages sent tothe same parent node may interfere with each other. By applyingFrequency DivisionMultiple Access (FDMA) or Code DivisionMultiple Access (CDMA), the schedule generated under com-petition is able to avoid collisions. However, the accompanyingmassive control packets will cause a large amount of energy tobe wasted. By using the control of BS, the energy waste can bereduced and thus the process may be much simpler. At the be-ginning of each round, the operation is also divided into severaltime slots. In the time slot, the node whose ID is i turns onits radio and receives the message from BS. BS uses the sameapproach to construct the routing tree in each round, and thenBS tells sensor nodes when to send or receive the data. In eachTDMA time slot, the nodes work in turns defined by BS. WhenBS receives all the data, the network will start the next phase.

D. Information Exchanging PhaseFor Case1, since each node needs to generate and transmit a

DATA_PKT in each round, it may exhaust its energy and die.The dying of any sensor node can influence the topography. Sothe nodes that are going to die need to inform others. The processis also divided into time slots. In each time slot, the nodes whoseenergy is going to be exhausted will compute a random delaywhich makes only one node broadcast in this time slot. Whenthe delay is ended, these nodes are trying to broadcast a packetto the whole network. While all other nodes are monitoring thechannel, they will receive this packet and perform an ID check.Then they modify their tables. If no such packet is received inthe time slot, the network will start the next round.For Case2, BS can collect the initial EL and coordinates

information of all the sensor nodes in Initial Phase. For eachround, BS builds the routing tree and the schedule of thenetwork by using the EL and coordinates information. Oncethe routing tree is built, the energy consumption of each sensornode in this round can be calculated by BS, thus the informationneeded for calculating the topology for the next round can beknown in advance. However, because WSN may be deployedin an unfriendly environment, the actual EL of each sensornode may be different from the EL calculated by BS. To copewith this problem, each sensor node calculates its EL anddetects its actual residual energy in each round. We define thecalculated EL as EL1 and the actual EL as EL2. When the two

Fig. 3. Routing tree generated by GSTEB for 100 nodes randomly deployed ina square for Case1.

ELs of a sensor node are different, the sensor node generates anerror flag and packs the information of actual residual energyinto DATA_PKT, which needs to be sent to BS. When thisDATA_PKT is received, BS will get the actual residual energyof this sensor node and use it to calculate the topology in thenext round.

V. COMPARATIVE ANALYSIS AND SIMULATION RESULTS

A MATLAB simulation of GSTEB is done for both Case1and Case2 to evaluate the performance.For Case1, we first compare GSTEB with PEGASIS and use

the same network model as PEGASIS. We generate a randomlydistributed 100 to 400 nodes network of square area m

m with BS located at (50 m, 175 m) and use DATA_PKTlength of 2000 bits and CTRL_PKT length of 100 bits. We leteach node have 0.25 J initial energy. Fig. 3 and Fig. 4 show therouting tree generated by GSTEB and PEGASIS for exactly thesame 100 node topology.In Fig. 3 the triangle is root node and in Fig. 4 the triangle is

head node and the rectangle is tail node. As seen, the routing treegenerated by GSTEB is better. Since PEGASIS uses GREEDYalgorithm to form a chain, long links may exist between parentnodes and child nodes, which will cause an unbalanced load.As for GSTEB, each node tends to choose the nearest neighbor.Even though it may not be able to select the global optimal so-lution, long links are avoided. Fig. 5 shows that the time whenthe first node dies changes within a range from 100 nodes to400 nodes in the network. Because network works in rounds,the time measurement unit is round but it is not a real time mea-surement unit. We can find that GSTEB performs much betterthan PEGASIS and prolongs network lifetime by about 100%to 300% in Fig. 5.To compare GSTEB with TBC, we use the same parameters

as TBC as shown in [17]. BS is located at (50 m, 120 m) and thelength of a DATA_PAK is 4000 bits. We compare the perfor-mance of GSTEB with the existing simulation results of TBC.


Fig. 4. Routing tree generated by PEGASIS for 100 nodes randomly deployedin a square for Case1.

Fig. 5. For Case1, we compare the time when first node dies for GSTEB andPEGASIS for the number of nodes from 100 to 400 in the square area, the net-work working in round makes Round be the time measurement unit.

Table I lists the lifetime of the sensor network in terms of theround when a node begins to die and in terms of the round whenall the nodes die.In Table I, the simulation results are taken from [17] except

GSTEB. We can find that GSTEB performs better than all otherprotocols. Both GSTEB and TBC are protocols which requirethat each node records the information of its neighbors and theyuse similar approach to transmit data. For GSTEB, each nodeneeds to record the information of its neighbors neighbors, soit needs more memory. But GSTEB can achieve a better perfor-mance in energy saving, because each node has more opportu-nities to choose the nearest neighbor as the parent. Moreover,there are several cluster-heads that need to communicate with

BS in TBC. The simulation results of TBC show that it performswell since BS is located near the sensor field. If BS is locatedfar away, the energy consumption of TBC increases quickly be-cause of the cluster-heads. And GSTEB will outperform it.When the network lifetime is defined as the time from the

start of the network operation to the death of the last node in thenetwork, PEDAP [8] achieves an optimal solution. It is becausePEDAP builds a minimum spanning tree, and it consumes theleast energy for data transmitting in each round. In this situation,we make some slight changes to GSTEB. We need not considerthe load balance, but in each round the total energy consumed bythe network should be the minimum. GSTEB chooses the nodenearest to BS as the root like PEDAP, and then GSTEB builds arouting tree and the topology remains unchanged until the rootdies. Once the root dies, the network reconstructs a new routingtree in the same way. We compare the improved GSTEB withPEDAP. Fig. 6 shows the routing tree generated by PEDAP ofthe same topology as Fig. 3 and Fig. 4. Simulation results showthat by using the same model as PEGASIS, in each round, thedata transmitting in the routing tree generated by the improvedGSTEB consumes less than 0.5% extra energy compared withthe routing tree generated by PEDAP. PEDAP is a protocol thatrequires BS to build the topography. To achieve that, BS shouldsend a broadcast packet which contains all the topology infor-mation and the schedule information to all sensor nodes or BSinforms the sensor nodes one by one when the topography needsto be rebuilt. If BS sends a packet which contains all the in-formation, each node has to receive the whole packet to get itsown information. This will cause a large amount of energy tobe wasted. On the other hand, with BS informing the nodes oneby one, the network can save energy, but causes a long delay.As compared, in the improved GSTEB, when the root node ischosen, all other nodes compute and find their own parents bythemselves in parallel without any information exchange, so theenergy consumption can be neglected. Once a node dies, othernodes remove its information from their tables and rebuild therouting tree. As a result, the improved GSTEB only consumesa little more energy for data transmitting than PEDAP, but theenergy for building the routing tree is greatly reduced, so theimproved GSTEB works almost as well as an optimal solution.For Case2, we change the initial energy of each node to 2 J

and J, we also generate a randomly distributed 100to 400 nodes network of square area m m with BSlocated at (50m, 175m) and use DATA_PKT length of 2000 bitsand CTRL_PKT length of 100 bits. We compare GSTEB withHEED in this case. Fig. 7 shows that the time when the first nodedies changes within a range from 100 nodes to 400 nodes in thenetwork. In Fig. 7 the timemeasurement unit is also round, but itis not a real timemeasurement unit. Fig. 8 shows the routing treegenerated by GSTEB for 100 nodes deployed in a square areain this case. Clearly, GSTEB performs far better than HEEDand prolongs the network lifetime by more than 100%. In Fig. 9and Fig. 10, we can find that almost all nodes have the sameresidual energy for GSTEBwhen the first node dies. For HEED,the nodes nearer to BS have much more residual energy than thenodes farther from BS. This is a typical load imbalance.HEED and GSTEB both aim to balance the network load for

Case2. HEED selects CHs by considering the residual energy,


Fig. 6. Routing tree generated by PEDAP for 100 nodes randomly deployed ina square for Case1.

Fig. 7. For Case1, we compare the time when first node dies for GSTEB andHEED for the number of nodes from 100 to 400 in the square area, the networkworking in round makes Round be the time measurement unit.

but only the local energy balance is attained. It is becauseCHs cannot dynamically decide to communicate with BS bysingle-hop or multi-hops. If they choose the single-hop, CHsfarther from BS consume more energy. On the other hand,even if they choose the multi-hops, CHs nearer to BS have totransmit more data and consume more energy. For GSTEB,each node dynamically decides to communicate with BS di-rectly or through others. All the nodes try to find neighborswith higher EL as parent nodes. In each round, the nodes withhigher EL consume more energy, which leads to a whole energybalance as shown in Fig. 9.Both in Case1 and Case2, GSTEB is suitable for different

initial energy circumstances. For Case1, nodes with the most

Fig. 8. Routing tree generated by GSTEB for 100 nodes randomly deployed ina square for Case2.

Fig. 9. Residual energy of the network when first node dies for GSTEB.

Fig. 10. Residual energy of the network when first node dies for HEED.

energy will be chosen as the root and communicate with BSfar away. In Case2, nodes with more energy will be selected as


parent more often. Both cases will cause the nodes with moreinitial energy to transmit or receive more data and consumemore energy.We calculate the waste of energy and the delay of GSTEB for

both cases.For Case1, since the topography is built by self-organizing,

each node is able to choose its parent simultaneously, it con-sumes little energy and this only causes a short delay in TreeConstructing Phase. In Self-Organized Data Collecting andTransmitting Phase, each leaf node needs to send a beaconwhich is received by its parent node. The beacon is a shortframe which contains only the ID of the leaf node. We assumethat the length of the beacon is 20 bits. Compared with theenergy consumption caused by DATA_PKT (2000 bits), theenergy waste caused by the beacon is only 1% of it. On theother hand, because all the leaf nodes transmit data at the sametime, the delay of data transmitting is j-based TDMA time slotswhere j is the maximal level of the tree. The delay is shorterthan that of PEGASIS and PEDAP.For Case2, BS should send a packet which contains the

topology information and the schedule information to the leafnodes one by one, which will lead to extra energy consumption.The CTRL_PKT is longer than the beacon in Case1 sinceit contains more information. We assume that the length ofCTRL_PKT sent by BS to each node is 200 bits. Each sensornode needs to receive a control packet in a round. Becausethe data cant be fused, it has to be transmitted for a longdistance, which causes significantly large energy consumption.Moreover, the relay nodes need to transmit more data, whichcauses a long delay. As a result, even though the process of theinformation exchanging between BS and sensor nodes leadsto a waste of energy and a delay, the ratio between wastefulconsumption and useful consumption is acceptable, and thismethod makes the process much simpler, so it is suitable forCase2.

VI. CONCLUSIONSIn this work, we introduce GSTEB. Two definitions of

network lifetime and two extreme cases of data fusion areproposed.The simulations show that when the data collected by sen-

sors is strongly correlative, GSTEB outperforms LEACH, PE-GASIS, TREEPSI [9] and TBC. Because GSTEB is a self-or-ganized protocol, it only consumes a small amount of energyin each round to change the topography for the purpose of bal-ancing the energy consumption. All the leaf nodes can transmitdata in the same TDMA time slot so that the transmitting delayis short. When lifetime is defined as the time from the start of thenetwork operation to the death of the first node in the network,GSTEB prolongs the lifetime by 100% to 300% compared withPEGASIS. In some cases, we are more interested in the lifetimeof the last node in the network. Some slight changes are made

to make the performance of GSTEB similar to that of PEDAP.So GSTEB is nearly the optimal solution in Case1.When the data collected by sensors cannot be fused, GSTEB

offers another simple approach to balancing the network load.In fact, it is difficult to distribute the load evenly on all nodesin such a case. Even though GSTEB needs BS to compute thetopography, which leads to an increase in energy waste and alonger delay, this kind of energy waste and longer delay areacceptable when compared with the energy consumption andthe time delay for data transmitting. Simulation results showthat when lifetime is defined as the time from the start of thenetwork operation to the death of the first node in the network,GSTEB prolongs the lifetime of the network bymore than 100%compared with HEED.

REFERENCES[1] K. Akkaya and M. Younis, A survey of routing protocols in wireless

sensor networks, Elsevier Ad Hoc Network J., vol. 3/3, pp. 325349,2005.

[2] I. F. Akyildiz et al., Wireless sensor networks: A survey, ComputerNetw., vol. 38, pp. 393422, Mar. 2002.

[3] Sohrabi et al., Protocols for self-organization of a wireless sensornetwork, IEEE Personal Commun., vol. 7, no. 5, pp. 1627, Oct. 2000.

[4] W. R. Heinzelman, A. Chandrakasan, and H. Balakrishnan, Energy-efficient communication protocols for wireless microsensor networks,in Proc. 33rd Hawaii Int. Conf. System Sci., Jan. 2000, pp. 30053014.

[5] W. B. Heinzelman, A. Chandrakasan, and H. Balakrishanan, An ap-plication-specific protocol architechture for wireless microsensor net-works, IEEE Trans. Wireless Commun, vol. 1, no. 4, pp. 660670, Oct.2002.

[6] O. Younis and S. Fahmy, HEED: A hybrid, energy-efficient, dis-tributed clustering approach for ad hoc sensor networks, IEEE Trans.Mobile Computing, vol. 3, no. 4, pp. 660669, 2004.

[7] S. Lindsey and C. Raghavendra, Pegasis: Power-efficient gatheringin sensor information systems, in Proc. IEEE Aerospace Conf., 2002,vol. 3, pp. 11251130.

[8] H. O. Tan and I. Korpeoglu, Power efficient data gathering and aggre-gation in wireless sensor networks, SIGMOD Rec., vol. 32, no. 4, pp.6671, 2003.

[9] S. S. Satapathy and N. Sarma, TREEPSI: Tree based energy efficientprotocol for sensor information, in Proc. IFIP Int. Conf., Apr. 2006,pp. 1113.

[10] I. F. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, Wire-less sensor networks: A survey, Computer Netw.s, vol. 38, no. 4, pp.393422, 2002.

[11] R. Szewczyk, J. Polastre, A.Mainwaring, and D. Culler, Lessons fromsensor network expedition, in Proc. 1st European Workshop on Wire-less Sensor Networks EWSN 04, Germany, Jan. 19-21, 2004.

[12] W. Liang and Y. Liu, Online data gathering for maximizing networklifetime in sensor networks, IEEE Trans Mobile Computing, vol. 6,no. 1, pp. 211, 2007.

[13] J. H. Chang and L. Tassiulas, Energy conserving routing in wirelessad hoc networks, in Proc. IEEE INFOCOM, 2000, vol. 1, pp. 2231.

[14] G. Mankar and S. T. Bodkhe, Traffic aware energy efficient routingprotocol, in Proc. 3rd ICECT, 2011, vol. 6, pp. 316320, .

[15] N. Tabassum, Q. E. K. Mamun, and Y. Urano, COSEN: A chain ori-ented sensor network for efficient data collection, inProc. IEEE ITCC,Apr. 2006, pp. 262267.

[16] M. Liu, J. Cao, G. Chen, and X. Wang, An energy-aware routing pro-tocol in wireless sensor networks, Sensors, vol. 9, pp. 445462, 2009.

[17] K. T. Kim and H. Y. Youn, Tree-Based Clustering(TBC) for energyefficient wireless sensor networks, in Proc. AINA 2010, 2010, pp.680685.

Documents

06782312.pdf