ieee sensors journal, vol. 16, no. 10, may 15, 2016 3981...

IEEE SENSORS JOURNAL, VOL. 16, NO. 10, MAY 15, 2016 3981

Fair Routing for Overlapped CooperativeHeterogeneous Wireless Sensor Networks

Kazuhiko Kinoshita, Member, IEEE, Natsuki Inoue, Yosuke Tanigawa, Member, IEEE,Hideki Tode, Member, IEEE, and Takashi Watanabe, Member, IEEE

Abstract— In recent years, as wireless sensor networks (WSNs)are widely diffused, multiple overlapping WSNs constructed onthe same area become more common. In such a situation, theirlifetime is expected to be extended by cooperative packet forward-ing. Although some researchers have studied about cooperation inmultiple WSNs, most of them do not consider the heterogeneityin the characteristics of each WSN such as battery capacity,operation start time, the number of nodes, nodes locations, energyconsumption, packet size and/or data transmission timing, and soon. In a heterogeneous environment, naive lifetime improvementwith cooperation may not be fair. In this paper, we propose a faircooperative routing method for heterogeneous overlapped WSNs.It introduces an energy pool to maintain the total amount ofenergy consumption by cooperative forwarding. The energy poolplays a role of broker for fair cooperation. Finally, simulationresults show the excellent performance of the proposed method.

Index Terms— Sensor network, cooperative routing, fairness,heterogeneous environment, load balancing.

I. INTRODUCTION

RECENTLY, wireless sensor networks (WSNs) havereceived much attention as a means for collecting and

utilizing data from real world. The number of WSN applica-tions has been increasing widely and the application range isexpected to spread [1], [2].

A WSN is a network composed of a large number ofsensor nodes with limited radio capabilities and one or afew sinks that collect data from sensor nodes. Generally,sensor nodes are powered by small batteries, hence, the energyconsumption in operating a WSN should be as low as possible.Some methods for prolonging network lifetime are required inWSNs [1]–[3].

Although all sensor nodes generate an equal amount ofdata packets in a WSN, nodes around a sink have to relaymore packets and tend to die earlier than other nodes becausethe energy consumption of sensor nodes is almost completelydominated by data communication rather than by sensingand processing. Hence, the whole network lifetime can be

Manuscript received January 7, 2016; revised February 25, 2016; acceptedFebruary 26, 2016. Date of publication March 8, 2016; date of current versionApril 8, 2016. This work was supported by the Strategic Information andCommunications Research and Development Promotion Program within theMinistry of Internal Affairs and Communications, Japan. The associate editorcoordinating the review of this paper and approving it for publication wasProf. Okyay Kaynak.

K. Kinoshita is with Tokushima University, Tokushima 770-8501, Japan(e-mail: [email protected]).

N. Inoue and T. Watanabe are with Osaka University, Osaka 565-0871,Japan (e-mail: [email protected]; [email protected]).

Y. Tanigawa and H. Tode are with Osaka Prefecture University,Osaka 599-8531, Japan (e-mail: [email protected]; [email protected]).

Digital Object Identifier 10.1109/JSEN.2016.2539360

prolonged by balancing the communication load at heavilyloaded nodes around a sink [4]. This issue is called the energyhole problem [5], [6] and is one of the most important issuesfor WSNs. There are numerous studies about load balancingfor WSNs such as clustering [7].

In addition, as WSNs are diffused widely, multiple over-lapping WSNs constructed on the same area become morecommon. In such a situation, cooperation among the WSNs toprolong network lifetime has been studied [8]–[10]. Assumingthat each sink of WSNs has a different location, the heavilyloaded area is also different. In this case, cooperation ofmultiple WSNs may be able to improve the network lifetimeof each WSN by load balancing all over the WSNs [11]–[13].

Note that even in a case where multiple WSNs are con-structed at the same place, they operate their applicationsindependently and they have heterogeneous characteristic fea-tures. However, most of the existing studies do not considerthis issue. For instance, if battery capabilities of sensor nodesin each network are different, in order to cooperate in aprofitable way, we need to consider some parameters, suchas their energy consumption rate, not only their remainingbattery. Otherwise, it is possible that certain WSNs prolongtheir lifetime but others shorten their lifetime. Since theirapplications are different, data sending interval and/or packetsize may be also different. Hence, for fair cooperation, it isnecessary to consider the total number of times that the nodehave forwarded a packet, instead of focusing on each packetforwarding only. Furthermore, operation start time, the numberof nodes and/or sensing area of each network may be alsodifferent.

In this paper, we consider the heterogeneity of networksand propose a fair cooperative routing method, to avoid unfairimprovement only on certain networks. We introduce oneor a few shared nodes that can use multiple channels torelay data packets. Assuming that sinks and shared nodescan communicate with any WSNs here, different WSNs canuse cooperative routing with each other since shared nodesallow sensor nodes to forward data from another WSN as thefunction of interchange points among respective WSN planes.When receiving a packet, a shared node selects the route tosend the packet, according to proposed route selection meth-ods. This cooperation prolongs the lifetime of each networkequally as possible.

II. RELATED WORK

A. Traditional Approach for Longer Lifetime

Clustering [7] is one of the most famous methods because ofits good scalability and the support for data aggregation. Data

1558-1748 © 2016 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.

3982 IEEE SENSORS JOURNAL, VOL. 16, NO. 10, MAY 15, 2016

aggregation combines data packets from multiple sensor nodesinto one data packet by eliminating redundant information.This reduces the transmission load and the total amountof data. In clustering, the energy load is well balanced bydynamic election of cluster heads (CHs) [14]. By rotating theCH role among all sensor nodes, each node tends to expendthe same amount of energy over time. Nevertheless, as withusual multihop forwarding, a CH around a sink tends to havehigher traffic than other CHs. As a result, nodes around sinksdie earlier than other nodes, even in clustered WSN [15].

In general, a single WSN has a single sink. The amount oftraffic increases around the sink, therefore nodes around thesink tend to die earlier. This is called energy hole problem.Moreover, in a large-scale WSN with a large number of sensornodes, the energy hole problem is more serious. Then, someresearchers have proposed construction methods of multiple-sink networks [16], [17]. In a multiple-sink WSN, sensor nodesare divided into a few clusters. Sensor nodes within a clusterare connected with one sink, which belongs to that cluster.In contrast to a single-sink WSN, in which nodes around thesink have to relay data from almost all nodes, nodes aroundeach sink relay smaller amount of data only from nodes thatare in the same cluster. Therefore, the communication load ofnodes around sinks can be reduced. However, there are someproblems such as how to determine the optimal location ofeach sink and the optimal number of sinks.

B. Cooperation Between Multiple WSNsIn existing studies, most researches assume that a single

network is deployed by a single authority in the sensing area.However, as WSNs get utilized more widely, multiple WSNstend to be deployed in the same area. For instance, in the UK,some different networks of cameras by different authoritiessuch as police, highway patrol, and local city authorities aredeployed on the same roads [18]. Recently, some researchershave proposed the cooperation method of multiple WSNs insuch situations.

When multiple WSNs are constructed in close prox-imity, they can help each other by forwarding data sothat all networks involved benefit from collaborative effort.In [11], the potential benefits of cooperation in multiple WSNsare investigated. The authors formulated the system modelwith objective function and a set of problem constraints.Then, a linear programming framework is used to solve theoptimization problem. Since their goal is to investigate themaximum achievable sensor network lifetime with differentmulti-domain cooperation strategies, optimization objective isnetwork lifetime, which is defined as the time when the firstsensor node in a network exhausts its battery and dies. Theauthors also investigated the cooperation in multiple networksthat are deployed slightly different location [12].

Some researchers have addressed the cooperation prob-lem with using a game-theoretic framework [20], [21]. It isassumed that a WSN has a rational and selfish character andwill only cooperate with another network if this associationprovides services that justify the cooperation.

Virtual Cooperation Bond (VCB) Protocol [20] is one ofthe game-theoretic approaches. It is a distributed protocol that

makes different networks to cooperate, if and only if all thenetworks obtain some benefits by the cooperation. The authorsformulated the cooperation problem among different WSNsas a cooperative game in game theory. In VCB protocol,the energy consumption of data communication is used ascosts. When the cost gets higher, the payoff of a network getslower. A sensor node and another node that belongs to anothernetwork forward a data packet coming from the other side,only if both networks can obtain the higher payoffs than nocooperation scenario. The simulation results showed that theVCB can save transmission energy between 20% and 30% ina certain environment.

C. Problematic Issues

As discussed above, we assume that multiple WSNs aredeployed by different authorities in the same area. ThoseWSNs operate different applications independently, hence,they have heterogeneous characteristics, such as batterycapacity, operation start time, the number of nodes, nodeslocations, energy consumption, packet size and/or data trans-mission timing. However, most existing cooperation methodsdo not consider this heterogeneity. For instance, when batteriescapacities on sensor nodes are quite different by a WSN, acooperative routing method based on residual energy is notappropriate since a WSN which has the maximum batterycapacity always forwards packets from other WSNs. As aresult, although certain WSNs prolong their lifetime, the otherWSNs may shorten their lifetime. In such a situation, fairnessof cooperation is a highly important issue.

In this paper, we aim to improve all WSNs lifetime by faircooperative routing in a heterogeneous environment, avoidingimproving the lifetime of only certain WSNs.

III. PROPOSED METHOD

A. Assumed Environment

In this paper, we assume the following environment.In a sensing field, m different WSNs are constructed, and

different applications are operating on each WSN indepen-dently. Figure 1 shows an example where two WSNs areconstructed. If heavy loaded nodes are in different placesamong the WSNs as indicated in the example, it is possiblethat data packets via heavy loaded nodes are forwarded byother nodes in another WSN. However, each network adoptsdifferent channel, hence sensor nodes are unable to commu-nicate with a node belonging to another WSN. To overcomethis limitation, q shared nodes, which are high-end nodes withmulti-channel communication unit, are deployed in the area.Shared nodes and sinks are able to communicate with anynodes belonging to all WSNs.

Sensor nodes consume their energy only by communication,which is a reasonable assumption in sensor networks withsimple sensors. Sinks and shared nodes have sufficiently largebatteries or power supply. We define the WSNs’ lifetime as thetime when a first sensor node depletes its all battery energy.

For heterogeneity, the battery capacity of a sensor node,the number of nodes, nodes’ locations, energy consumptionby communication, packet size, data transmission timing and

KINOSHITA et al.: FAIR ROUTING FOR OVERLAPPED COOPERATIVE HETEROGENEOUS WSNs 3983

Fig. 1. Two WSNs deployed at the same area.

operation start time are different by each WSN. Note that thesensing area is the same in all WSNs since we aim at thecooperation in overlapped multiple networks.

B. System Model

In this subsection, we formulate the overlapped WSNsmodel for fair cooperation routing.

In a sensing field, m different WSNs N1, · · · , Nm areconstructed, and each network Ni , 1 ≤ i ≤ m, has a setof unique sensor nodes Ni = {ni1, ni2, . . . , ni|Ni |} and thesink BSi . q shared nodes s1, . . . , sq also exists in the area.All WSNs are able to use these shared nodes as relay nodefor packet forwarding.

For guaranteeing the lifetime improvement by the cooper-ation, we define network lifetime Li , the estimated lifetimeof Ni , is obtained by Eq. (1).

Li = minni j ∈Ni

Li j (1 ≤ j ≤ |Ni |). (1)

Li j is the estimated lifetime of the sensor node ni j here. Wecall it node lifetime. In other words, the estimated lifetimeof a WSN is a minimum estimated lifetime of its all sensornodes. Each sensor node measures its own energy consumptionduring specific time τ and calculates Li j by using it. Let ei j t

be the remaining energy of node ni j at time t , then, energyconsumption per unit time is described by

ei j t − ei j (t+τ )

τ, (2)

and Li j is represented by Eq. (3).

Li j = ei j (t+τ ) · τ

ei j t − ei j (t+τ )(3)

By exchanging Li j periodically among neighboring nodes,each node updates Li . In addition, minimum lifetime L0

i , theestimated lifetime in the case of no cooperation, is calculatedby each sensor node. Specifically, each WSN operates withoutany cooperation from time t = 0 to t = 0 + τ = τ , and afterthe duration, L0

i is calculated by Eq. (4).

L0i = ei jτ · τ

ei j0 − ei jτ(4)

L0i is also exchanged and updated among sensor nodes.

Specific updating procedure of Li and L0i is explained in

Sect. 3.4.

A shared node sk (1 ≤ k ≤ q), has m routes Rikl to the sink

BSi via network Nl (1 ≤ l ≤ m). Hence, sk selects one ofthe m routes when sk receives a data packet from network Ni .If i �= l, Ni rents the energy resource from Nl .

Moreover, we define route lifetime L Rikl

as the estimated

lifetime of the route Rikl . The detailed definition is as follows.

L Rikl

= minnl j ∈Ri

kl

Ll j (5)

Eq. (5) means that L Rikl

is the minimum lifetime among the

nodes being contained in route Rikl .

C. Route Discovery

Each sensor node creates its routing table based on a routingprotocol. In this paper, we used ad hoc on-demand distancevector (AODV) [19] as a routing protocol, because AODVwas developed for wireless ad hoc networks and was adoptedfor some WSN protocols such as Zigbee [22] and ANT [23].In route discovery, each sensor node discovers its routes notonly to the sink in its WSN but also to all the other sinks inthe different WSNs for opportunities to forward data packetsfrom nodes in different WSNs to their sink. Therefore, therouting table of each sensor node has m routes correspondingto each sink in all WSNs.

A shared node discovers its route with a slightly differentmechanism. A shared node creates m routes via m differentWSNs to a sink. There are m sinks, in total, corresponding tom WSNs. Therefore, a shared node has m × m routes.

In AODV route discovery, each node chooses a route thathas the minimum number of hops to the sink. However, theproposed method uses not the number of hops but a costcalculated by simple accumulation, so that more routes areestablished via shared nodes. This is because different WSNscan be used only via shared nodes as alternative routes.Specifically, we set 1 as the cost of going through a sensornode and we set x(0 < x < 1) as the cost of going througha shared node. When each node discovers a route, it choosesa route that has the minimum cost calculated as the sum oftraversing nodes. Another advantage of the proposed routediscovery is that using shared nodes, which have sufficientlylarge batteries or power supply, is expected to reduce powerconsumption of other sensor nodes.

D. Obtaining Lifetime Information

For cooperation considering the fairness among multipleWSNs, shared node sk maintains estimated lifetime infor-mation, network lifetime Li , minimum lifetime L0

i and routelifetime L Ri

kl. We explain how to obtain these information as

follows.At the time of transmitting a data packet, sensor node

ni j adds the values of its network lifetime Li and routelifetime L Ri

klto the MAC frame header of the packet. If the

node does not have any information on network lifetime orroute lifetime yet, for instance at the time immediately aftercreating or updating the route, its own node lifetime Li j isadded alternatively. Each node updates these information by


Fig. 2. Obtaining lifetime information.

overhearing data packets from other nodes. Specifically, whennode ni j overhears a data packet, it compares the value ofthe network lifetime in the data packet and Li in its owninformation, and updates its own Li to the smaller valuebetween them. In addition, if the packet is from a nodewhich is contained in Ri

j i , the route from ni j to BSi , itchecks the value of route lifetime in the packet header, andupdates its route lifetime by the smaller value as in the caseof updating Li . After that, the overhearing node discards thepacket immediately if the destination of the packet is notitself. As we mentioned in Section 3.2, network lifetime for thetime 0 to τ is represented as minimum lifetime L0

i . To obtainthis value, each node updates its minimum lifetime with thevalue of network lifetime on an overheard packet, from thetime τ to 2τ . Figure 2 describes this mechanism to obtainlifetime information.

E. Cooperative Data Forwarding

Since a sensor node has a single route toward the sink inits WSN, it forwards a data packet immediately to the nextnode on the route. On the other hand, a shared node sk has mroutes for the sink via m networks, therefore it can choose anappropriate route for data forwarding.

Since the lifetime of WSN depends on the lifetime of theenergy-bottleneck nodes in the WSN, cooperative data packetforwarding via alternative nodes belonging to another WSNinstead of the bottleneck nodes is expected to improve the

Fig. 3. Example of the cooperative routing with a shared node intwo WSNs.

lifetime of the WSN. An example is described in Figure 3.Here, the sensor nodes of WSN 2 between the shared nodeand sink 1 can forward data packets to sink 1 for WSN 1as an alternative route on another WSN. However, if thealternative nodes are also bottleneck of their WSN, the lifetimeof their WSN would be shortened. To avoid this result, ashared node is able to choose the alternative route only if thealternative nodes are not bottleneck. That is, the condition thatpacket forwarding from the shared node sk to the sink BSi inWSN i via route Ri

kl of WSN l is available can be formulatedas follows.

L Rikl

> L0l (6)

By this condition, lifetime reduction of each WSN by forward-ing packets from other WSNs is avoided, and the improvementof WSNs lifetime is guaranteed.

As explained in Section 3.3, a shared node has multipleroutes to the sink, hence an algorithm to select an appropriateroute is needed. We propose fair cooperative methods with tworoute selecting algorithms. The first one is named Pool-basedselecting. We resemble the cooperative forwarding to debtof energy resource. Shared nodes maintain the Energy Pool,the total amount of energy consumption used by cooperativeforwarding, continuously. When a node nl j in Nl forwards apacket from another network Nu , the Energy-Pool of Nl isincreased and that of Nu is decreased. By selecting a routebased on the value of Energy-Pool, the cooperation with thefairness of energy consumption is achieved in a heterogeneousenvironment. In addition, this method is able to balance theenergy consumption by cooperation even if each WSN startsto operate from different time.

When a shared node sk receives a data packet, if it hasmultiple available routes to the sink, it compares the Energy-Pool Pl of each network Nl , and selects the route that hasminimum P . Let Ri

kv denote the selected route from sharednode sk to sink BSi via Nv . Pv , the Energy-Pool of the networkNv which the Ri

kv belongs to is increased, and Pi , the Energy-Pool of the network Ni which the source node belongs to isdecreased. The amount of increase and decrease �P is theenergy consumption by packet forwarding from sk to BSi .It is obtained by Eq. (7).

�P = h Rikv

(Erv + Etv) (7)


Fig. 4. Pool-based route selection.

Fig. 5. Life-based route selection.

h Rikv

is the hop count on Rikv . Erv and Etv are reception and

transmission energy cost per a packet on Nv , respectively.Figure 4 shows the flow chart on Pool-based route selectionat the time when a shared node sk receives a data packetfrom Na .

The other is named Life-based selecting, that selects theroute with maximum route lifetime. In contrast to the Energy-based route selection that considers only remaining energyon the nodes, Life-based is focusing on the traffic loads byestimating the route lifetime. Therefore, it is expected that theheavy-loaded nodes balance their loads to other network nodesand it leads to a longer lifetime. Figure 5 shows the procedureof the Life-based route selection.

IV. PERFORMANCE EVALUATION

A. Simulation Environment

We evaluated the performance of the proposed method withthe network simulator QualNet 7.1 [24]. We observed thereceiving rate, which is the rate of sensor nodes that send datapackets to their sinks successfully. Therefore, we counted anode that cannot communicate with its sink as a dead node, inspite of its remaining battery. The maximum value of receivingrate is 1.

In this simulation model, we set the node configurationsusing datasheet and information provided by MEMSIC [25].We simulated four WSNs, WSN 1, WSN 2, WSN 3 and WSN4as follows. Each WSN had 49 nodes based on a randomtopology. The sensing field was a 490 m ×490 m square.

Fig. 6. Receiving rate on WSN 1.

The PHY model was IEEE802.11b and its data rate was2 Mbps. The maximum range of radio transmission for eachnode was 150 m.

Each sink was located at each corner of the field.A shared node was placed at the center of the field. Eachnode sent 512 bytes data packets asynchronously at intervalsof 10 seconds. We assumed that sinks and shared nodes had asufficiently large battery, and that their battery capacities wereunlimited. We set x , the cost of using a shared node, to 0.5.To give opportunities for cooperative forwarding to sensornodes fairly, all nodes deleted their route entries and discov-ered new routes at intervals of 720 minutes.

We evaluated two proposed method, Pool-based and Life-based. For comparison, we simulated an environment wherefour WSNs were operated independently without any cooper-ation. In addition, Energy-based method was also evaluatedas a conventional method. It just focuses on prolonging totallifetime but ignores the fairness among WSNs.

B. Simulation Results

1) Scenario 1: Heterogeneous Battery Capacity: As a basicevaluation for heterogeneity, sensor nodes have different bat-tery capacity by a WSN. WSN 1 has the largest capacityand WSN 4 has the lowest. We set the battery capacity of anode in WSN 1 to 1, and the capacity ratio is represented as;W SN1 : W SN2 : W SN3 : W SN4 = 1 : 0.75 : 0.625 : 0.5.Note that each node does NOT need to know the initialcapacity of nodes in other WSNs. All each node has to knowis its own initial capacity for operating the proposed methodproperly.

Figures 6-9 show the receiving rate as a function of elapsedtime for each WSN. They are averaged over 10 trials. We cansee that Pool-based and Life-based cooperation extend thelifetime of W SN1 in Fig. 6. Especially, Pool-based achieveddramatic improvement. On the other hand, Energy-basedcooperation degraded the lifetime of W SN1, since W SN1has larger battery capacity than any other WSNs. On Energy-based, a shared node always selects the route that has the max-imum residual energy. Therefore, in this scenario, the routevia W SN1 forwarded a lot of packets from other WSNs andW SN1 consumed much more energy than any other WSNs.





As a result, the lifetime of W SN1 was shortened. To thecontrary, on Pool-based, the route via each WSN forwarded thealmost same amount of data packets. Hence, W SN1 was alsoable to improve its lifetime. On Life-based, since a shared nodecompares the estimated lifetime of the routes, the heavy-loadednodes tend to be avoided even if they belongs to W SN1.In Figs. 7, 8 and 9, we can see both the proposed methodand conventional method improved network lifetime.

For evaluation, we define the α−li f etime as the time whenthe receiving rate has fallen below α on a WSN. We also definelife improving ratio, which is represented by α − li f etime

Fig. 10. Average of life improving ratio in 4 WSNs (scenario 1).

Fig. 11. Total amount of extended lifetime in 4 WSNs (scenario 1).

on the method divided by α − li f etime in no cooperationscenario.

Figure 10 shows the average life improving ratio in 4 WSNsfor each method as a function of α. All methods extended thenetwork lifetime by cooperative forwardings. In most of otherrange than α close to 1, specifically, Life-based and Pool-basedachieved greater benefits than Energy-based.

Since the networks have different battery capacities, thelifetime of them without cooperation are also different. Evenif the total amount of extended lifetime is equal, the lifeimproving ratio may take a larger value with smaller batterycapacity. Hence, for confirmation, we present the total amountof extended lifetime in 4 WSNs in Figure 11. We can seeslightly different behavior from Figure 10. Pool-based methodachieved the maximum improvement in all range, since itcooperated more aggressively than Life-based method.

In addition, Figure 12 shows the variance of life improv-ing ratio in 4 WSNs for each method. Obviously, the vari-ance on Pool-based is remarkably small. This result impliesPool-based method achieved fair cooperation in a heteroge-neous environment. The energy-pool successfully plays a roleof broker for cooperation.

2) Scenario 2: Heterogeneous Data Transmission: We eval-uated the 4 WSNs that send data packets in different timing.In this scenario, WSN1, WSN2, WSN3 and WSN4 senda data packet every 10 minutes, every 7.5 minutes,


Fig. 12. Variance of life improving ratio in 4 WSNs (scenario 1).

Fig. 13. Average of lifetime improving ratio in 4 WSNs (scenario 3).

Fig. 14. Variance of lifetime improving ratio in 4 WSNs (scenario 3).

every 6.25 minutes and every 5 minutes, respectively. Thesevalues were not special. In this scenario, we intended toevaluate how the proposed method works in a case where eachWSN collects data in deferent timings. In other words, a WSNwith larger interval consumes its battery more slowly and mayhave to forward more packets from other WSNs unfairly. Otherparameters are the same as scenario 1 except that all sensornodes in any WSNs have the equal battery capacity.

We do not present any graphs for the life improve ratioin scenario 2 since the results are very similar to scenario 1.We observed that the proposed methods extended lifetime of

all WSNs fairly. Due to the non-uniform traffic modeling, par-ticular areas may get congested temporarily. But, the assumedpacket generation interval is long enough, so that collisionscan be avoided by CSMA/CA manner.

3) Scenario 3: Heterogeneous Operation Start Time: In thisscenario, each WSN starts its operation at different time.WSN1, WSN2, WSN3 and WSN4 start to work at 0, 1000,2000 and 3000 minutes, respectively. Other parameters are thesame as scenarios 1 and 2, with the same battery capacity andthe same data sending interval.

Figure 13 shows the averaged lifetime improving ratio over4 WSNs in scenario 3. Pool-based cooperation achieved themaximum lifetime improvement.

Moreover, in Figure 14, we can see that the proposedmethods obtained quite smaller variance than the conven-tional method also in scenario 3. Note that the variance ofPool-based cooperation is slightly larger than in scenario 1,since a network that started operating at earlier time has moreopportunities to cooperate than others. We can see this fact inFigure 14.

V. CONCLUSIONS

In this paper, we focused on heterogeneous overlappedsensor networks that were constructed at the same area.In such a situation, it is expected that the lifetime of allnetworks should be extended by cooperation in multiple net-works. However, since the existing methods do not considerthe heterogeneity in each network, fairness in terms of lifetimeimprovement is required. We proposed a fair cooperativerouting method with shared nodes, with the aim to achievefair lifetime improvement in heterogeneous overlapped sensornetworks.

Simulation results showed that the proposed methodextended the network lifetime. In particular, Pool-based coop-eration achieved quite small variance of lifetime improvement,that is, it provided quite fair cooperation.

As a future work, we try to implement the proposed methodon an experimental system and evaluate its feasibility.

REFERENCES

[1] J. Yick, B. Mukherjee, and D. Ghosal, “Wireless sensor network survey,”Comput. Netw., vol. 52, no. 12, pp. 2292–2330, Aug. 2008.

[2] I. F. Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, “A surveyon sensor networks,” IEEE Commun. Mag., vol. 40, no. 8, pp. 102–114,Aug. 2002.

[3] I. Dietrich and F. Dressler, “On the lifetime of wireless sensor networks,”ACM Trans. Sensor Netw., vol. 5, no. 1, Feb. 2009, Art. no. 5.

[4] M. Perillo, Z. Cheng, and W. Heinzelman, “On the problem of unbal-anced load distribution in wireless sensor networks,” in Proc. IEEEGLOBECOM Workshops Wireless Ad Hoc Sensor Netw., Dec. 2004,pp. 74–79.

[5] J. Li and P. Mohapatra, “Analytical modeling and mitigation techniquesfor the energy hole problem in sensor networks,” Pervasive MobileComput. J., vol. 3, no. 3, pp. 233–254, Jun. 2007.

[6] X. Wu, G. Chen, and S. K. Das, “Avoiding energy holes in wireless sen-sor networks with nonuniform node distribution,” IEEE Trans. ParallelDistrib. Syst., vol. 19, no. 5, pp. 710–720, May 2008.

[7] A. A. Abbasi and M. Younis, “A survey on clustering algorithms forwireless sensor networks,” Comput. Commun., vol. 30, nos. 14–15,pp. 2826–2841, Oct. 2007.

[8] L. Buttyán, T. Holczer, and P. Schaffer, “Spontaneous cooperation inmulti-domain sensor networks,” in Proc. 2nd Eur. Workshop Secur.Privacy Ad-Hoc Sensor Netw., Jul. 2005, pp. 42–53.


[9] E. De Poorter, B. Latré, I. Moerman, and P. Demeester, “Symbiotic net-works: Towards a new level of cooperation between wireless networks,”Int. J. Wireless Pers. Commun., Jun. 2008, pp. 479–495.

[10] J. Steffan, L. Fiege, M. Cilia, and A. Buchmann, “Towards multi-purpose wireless sensor networks,” in Proc. Int. Conf. SensorNetw. (IEEE SENET), Aug. 2005, pp. 336–341.

[11] K. Bicakci and B. Tavli, “Prolonging network lifetime with multi-domaincooperation strategies in wireless sensor networks,” Ad Hoc Netw., vol. 8,no. 6, pp. 582–596, Aug. 2010.

[12] K. Bicakci, I. E. Bagci, B. Tavli, and Z. Pala, “Neighbor sensornetworks: Increasing lifetime and eliminating partitioning through coop-eration,” Comput. Standards Interfaces, vol. 35, no. 4, pp. 396–402,Jun. 2013.

[13] J. Nagata, Y. Tanigawa, K. Kinoshita, H. Tode, and K. Murakami,“A routing method for cooperative forwarding in multiple wireless sen-sor networks,” in Proc. 8th Int. Conf. Netw. Services (ICNS), Mar. 2012,pp. 43–46.

[14] G. Gupta and M. Younis, “Performance evaluation of load-balancedclustering of wireless sensor networks,” in Proc. 10th Int. Conf.Telecommun. (ICT), Mar. 2003, pp. 1577–1583.

[15] X. Du, Y. Xiao, and F. Dai, “Increasing network lifetime by balancingnode energy consumption in heterogeneous sensor networks,” WirelessCommun. Mobile Comput., vol. 8, no. 1, pp. 125–136, Jan. 2006.

[16] E. I. Oyman and C. Ersoy, “Multiple sink network design problemin large scale wireless sensor networks,” in Proc. IEEE Int. Conf.Commun. (ICC), vol. 6, Jun. 2004, pp. 3663–3667.

[17] F. Fabbri, C. Buratti, and R. Verdone, “A multi-sink multi-hop wirelesssensor network over a square region: Connectivity and energy consump-tion issues,” in Proc. IEEE GLOBECOM Workshops, Nov./Dec. 2008,pp. 1–6.

[18] C. Efstratiou, I. Leontiadis, C. Mascolo, and J. Crowcroft, “Demoabstract: A shared sensor network infrastructure,” in Proc. 8th ACMConf. Embedded Netw. Sensor Syst. (SenSys), 2010, pp. 367–368.

[19] C. E. Perkins, E. M. Royer, and S. R. Das. (Nov. 2001). Ad HocOn-Demand Distance Vector (AODV) Routing. [Online]. Available:http://tools.ietf.org/html.draft-ietf-manet-aodv-09.

[20] P. O. S. Vaz de Melo, F. D. Cunha, and A. A. F. Loureiro, “A distributedprotocol for cooperation among different wireless sensor networks,” inProc. IEEE Int. Conf. Commun. (ICC), Jun. 2013, pp. 6035–6039.

[21] M. J. Shamani, H. Gharaee, S. Sadri, and F. Rezaei, “Adaptiveenergy aware cooperation strategy in heterogeneous multi-domainsensor networks,” Procedia Comput. Sci., vol. 19, pp. 1047–1052,2013. [Online]. Available: http://www.sciencedirect.com/science/article/pii/S1877050913007552?np=y

[22] ZigBee Alliance, accessed on Dec. 28, 2015. [Online]. Available:http://www.zigbee.org

[23] This is Ant, accessed on Dec. 28, 2015. [Online]. Available:http://www.thisisant.com/

[24] QualNet Simulator Version 7.1, accessed on Dec. 28, 2015. [Online].Available: http://web.scalable-networks.com

[25] MEMSIC: MICAz Datasheet, accessed on Dec. 28, 2015. [Online].Available: http://www.memsic.com/userfiles/files/Datasheets/WSN/micaz_datasheet-t.pdf

Kazuhiko Kinoshita (M’99) received the B.E., M.E., and Ph.D. degreesfrom Osaka University, Osaka, Japan, in 1996, 1997, and 2003, respectively,all in information systems engineering. From 1998 to 2002, he was anAssistant Professor with the Department of Information Systems Engineering,Graduate School of Engineering, Osaka University. From 2002 to 2008, hewas an Assistant Professor with the Department of Information Networking,Graduate School of Information Science and Technology, Osaka University,where he was an Associate Professor from 2008 to 2015. Since 2015, he hasbeen a Professor with the University of Tokushima. His research interestsinclude mobile networks, network management, and agent communications.He is a Senior Member of the Institute of Electronics, Information andCommunication Engineers.

Natsuki Inoue received the bachelor’s degree in engineering and master’sdegree in information networking from Osaka University, Osaka, Japan, in2012 and 2015, respectively. Since 2015, he has been with Rakuten, Inc.

Yosuke Tanigawa (M’09) received the B.E., M.E., and Ph.D. degrees fromOsaka University, in 2005, 2007, and 2009, respectively, all in informationnetworking. He is currently an Assistant Professor with the Department ofComputer Science and Intelligent Systems, Graduate School of Engineering,Osaka Prefecture University. His current research interests include multime-dia communications and QoS-aware network controls in wireless networks.He is a member of The Institute of Electronics, Information and Communi-cation Engineers, Japan.

Hideki Tode (M’92) received the B.E., M.E., and Ph.D. degrees from OsakaUniversity, in 1988, 1990, and 1997, respectively, all in communicationsengineering. He was adopted as an Assistant Professor with the Departmentof Communications Engineering, Osaka University, from a Ph.D. Student in1991. In 1998 and 1999, he shifted and promoted to Lecturer and AssociateProfessor with the Department of Information Systems Engineering, andshifted to the Department of Information Networking, Graduate School ofInformation Science and Technology, Osaka University, in 2002. Since 2008,he has been a Professor with the Department of Computer Science and Intel-ligent Systems, Graduate School of Engineering, Osaka Prefecture University.His current research interests include architectures of optical network, wirelessmultihop network, future Internet, and contents distribution technologies.He is a Senior Member of The Institute of Electronics, Information andCommunication Engineers, Japan.

Takashi Watanabe (M’87) received the B.E., M.E., and Ph.D. degreesfrom Osaka University, Japan, in 1982, 1984, and 1987, respectively. Hejoined Faculty of Engineering, Tokushima University, in 1987, as an AssistantProfessor, and moved to the Faculty of Engineering, Shizuoka University, in1990. He was a Visiting Researcher with the University of California, Irvine,from 1995 to 1996. He is currently a Professor with the Graduate School ofInformation Science and Technology, Osaka University, Japan. His researchinterests include wireless networking, mobile networking, ad hoc networks,sensor networks, ubiquitous networks, intelligent transport systems, especiallyMAC and routing. He is a member of the IEEE Communications Society, theIEEE Computer Society, the Information Processing Society of Japan (IPSJ),and the Institute of Electronics, Information and Communication Engineers,Japan (IEICE). He has served on many program committees for networkingconferences, such as the IEEE, the Association for Computing Machinery,IPSJ, and IEICE.

ieee sensors journal, vol. 16, no. 10, may 15, 2016 3981...

Documents