Enhanced Fuzzy Based Clustering Approach for ...

Proceedings of International Conference on Advances in Applied Engineering and Technology – 2015, May 14-16, 2015.

Enhanced Fuzzy Based Clustering Approach for Improving Reliability of WSNs 1

Gurumekala T, 2Senthil Sivakumar M, 3Sundaram A, 4ArputhaRaj T

1

Departmentof Computer Science Engineering,PSR Engineering College, Sivakasi, India. 2 DepartmentofElectronics, Madras Institute of Technology,Chennai,India. 3 Department of Electronics,Wolkite university, Wolkite, Ethiopia. 4 School of Electronics and Communication,St.Joseph university in Tanzania, East Africa. Abstract Wireless Sensor Network acts as a bridge between real physical and virtual world and also plays a vital role in various applications which are associated with capturing environment status. Energy constraint is a major problem in such applications that must be considered. Wireless sensor networks (WSNs) are partitioned into clusters for the purpose of efficient information collection. Clustering method is capable of making an effective way to extend the lifetime of WSNs. Current clustering methods don’t provide effective cluster head selection which leads to reduction of network lifetime. To combat such a critical situation, this approach takes into account expected residual energy and distance to the base station of a sensor node. Expected Residual Energy (ERE) is stated as the predicated remaining energy to be selected as a cluster head. The nodes with maximum ERE are periodically rotated to allocate the workload evenly so as to conserve energy of the nodes. The result preciously exposes that this approach outperforms the existing methods and has been proven its efficiency. Keywords:Energy Predication, Expected Residual Energy, Cluster Head Selection, Wireless Sensor Networks, Fuzzy interference System. 1. Introduction The tremendous advances in the semiconductor manufacturing technology, digital electronics and wireless communications during the last decades leading to the development of low-cost, low-power, and small-sized devices with computing, embedded sensing, and communication capabilities. A WSN is a masscollection of such sensor nodes which carry out distributed sensing tasks [1]-[6] where RF (radio frequencies) are used. The severe design issue of WSNs is energy efficiency as the sensor nodes are equipped with nonrechargeable batteries that must be taken into account to extend the life-time of sensor networks. Cluster-based design is one of the ways to preserve the energy of sensor nodes since some sensor nodes which are less in number, called Cluster Heads (CHs) are allowed to communicate with the base station. The important work of CHs are, collecting the data sent by each node to the cluster, compress it, and then transmit the data after aggregation to the base station. The representative design to cluster the sensor networks called LEACH (low-energy adaptive clustering hierarchy) method purely utilizes the probabilistic model. The Cluster Head (CH) selection is done based on the probabilistic model and the CHs are periodically rotated to balance energy consumption. Even though LEACH exhibits its better

performance, in some situations, possibilities are there to select inefficient CHs. Since LEACH fully depends on only a probabilistic model, there may be possibilities to select the nodes close each other and can be located as CHs in the edge of WSNs. These cluster heads are not capable of maximizing the energy efficiency of the sensor networks. Appropriate cluster-head selection can significantly minimize the energy consumption and prolong the lifetime of sensor networks. Some of the clustering algorithms aimed to design WSN use to manage uncertainties in WSN. In general, fuzzy based clustering algorithms to choose cluster heads. Gupta et al. [9] introduced a method to escape from the problems associated with LEACH which utilizes three fuzzy descriptors called residual energy, concentration, and centrality in the course of the cluster-head selection had been introduced by. The concentration indicates the number of nodes present in the vicinity; the centrality denotes a value which classifies the nodes based on how central the node is to the cluster. The selection of CHs are centrally done at a base station. All sensor nodes transmit its clustering information to that base station. Since this mechanism is a centralized approach, each CHs are overwhelmed. Kim et al. [10] proposed an analogous approach namely Cluster Head Election Mechanism using Fuzzy Logic (CHEF) with the use of two fuzzy descriptors (Residual Energy and Local Distance). The local distance is the total distance between the hesitant CH and the nodes within predefined constant competition radius. Therefore, base station is not necessary to gather clustering information from all sensor nodes. In general, different environments with different characteristicswon’t make the process of electing CHs easy. For that Anno et al. [11] planned to have different fuzzy descriptors called number of neighbor nodes, counting the remaining battery power of a node, distance from the cluster centroid, and the network traffics, and assessed their performance. The sensor nodes which are neighbor to the base station drink much more energy because of increased network traffic nearby the base station. Therefore, the neighbor of base station quickly loses their battery power. Apart from the residual energy, Bagci et al [12] further introduced a fuzzy descriptor namely distance to the base station, during the selection of cluster head. This irregular clustering approach is completely based on the knowledge of decreasing the cluster sizes when getting closer to the base station. All existing clustering approaches considered the residual energy of sensor nodes while CH selection. Nevertheless, the remaining energy of CH after being selected and running around has never been discussed. A round refers to the intermission between two consecutive cluster formation processes. In this work, a new fuzzy based clustering approach with three fuzzy

Organized by Syed Ammal Engineering College, Ramanathapuram, Tamilnadu, India.


descriptors namely residual energy, Expected Residual Energy(ERE) and network traffic in addition to that, energy predication has been proposed to extend the lifetime of WSNs by evenly allocating the work assignment during the selection of CHs. The expected energy consumption (EEC) is a compulsory attribute to calculate the ERE. In our work, the EEC would be quickly assessed through an off-line trained neural network model. The proposed work follows LEACH architecture with an extension to the energy predication based on the ERE, and therefore the approach is named as LEACH-ERE. Expected/Estimated Remaining Energy is utilized in clusterhead selection for the first time in WMNs’ clustering approach. The rest of this paper is systematized as follows. Section 2 concisely introduces the scheme for predication of energy consumption. Next, proposed novel method is explained in Section 3. Then, the simulation example with 100- sensor node WSN has been used to assess the performance of proposed scheme. Finally, Section 5 contributes conclusion of this paper. 2. LiteratureReview Predication of Energy Consumption - LEACH Clustering algorithm: Low Energy Adaptive Clustering Hierarchy (LEACH) is the first hierarchical cluster-based routing protocol for wireless sensor network which partitions the nodes into clusters, in each cluster a reserved node with extra privileges called Cluster Head (CH) is responsible for creating and manipulating a TDMA (Time division multiple access) schedule and sending aggregated data from nodes to the BS where these data is needed using CDMA (Code division multiple access). Rests of the nodes are cluster members. This clustering protocol is divided into two rounds; each round consists of two phases, such as Set-up Phase: consists of Advertisement Phase and Cluster Set-up Phase; Steady Phase: consists of Schedule Creation and Data Transmission. Even though LEACH protocol shows its good manner, it gets suffered from many drawbacks which are listed below: CH selection is done randomly, that does not take into account energy consumption, can't cover a large area, CHs are not evenly distributed, where there is a possibility to locate CHs at the edges of the cluster. As LEACH has many demerits, many researchers have been strived hard to make this protocol performs better. LEACH forms clusters with the use of distributed algorithm, where nodes perform autonomous decisions in noncentralized way. Each sensor node i takes part to elect itself to be a CH at the beginning of round r+ 1 (which starts at time t) with probability Pi(t). Pi(t)is chosen such that the expected number of CHs for this round is k. If the networkconsists of N nodes, each node would choose to become a CH at round r with the probability as (1). (1) Where Ci(t) is the indicator function determining whether or not node i has been a CH within the most recent (r mod N /K ) rounds (Ci(t)= 0 means node i has been a CH). Thus, only nodes that have not already been CHs recently (i.e. Ci(t)= 1) may become CHs at round r+ 1. Radio Energy Dissipation Model: Currently, there is a great deal of research in the area of

low-energy radios. In this paper, the first-order radio model shown in [16] has been adopted to model the energy dissipation.

Fig.1: Formation of Clusters and Operation As the distance between the transmitter and receiver is less than a threshold value d0, the free space model (d2 power loss) is employed. Otherwise the multipath fading channel model (d4 power loss) is used. Equation (2) denotes the amount of energy consumed for transmitting l bits of data to d distance, while (3) represents the amount of energy consumed for receiving l bits of data. (2) (3) EelecTxand

EelecRx are the terms used for energy consumption per bit in the transmitter and receiver circuits. Also, εmp are the energy consumption factor of amplification for the free space and multipath radio models, respectively. The threshold value d0 could be obtained via (4). (4) Expected Residual Energy: Before the cluster formation, the number of cluster members is unknown. However, since it is proportional to the number of neighbors near a potential CH (located in a specific transmission range), the number of neighbors (stated as value n) could be used to obtain the expected energy consumption during the CH selection. As shown in Fig. 1, after the cluster formation, the steady-state operation is broken down into frames, where nodes transmit their data to the CH at most once per frame during their allocated transmission slot. In a frame, suppose a CH has n cluster members, it would receive n messages from all the members and then transmit one combined message to the base station with a distance dtoBS. The number of frames could be acquired by (5). (5) wheretssPhase is the operation time of the steady-state phase (i.e. the time of a node to be a CH), tslot denotes the slotted time necessary for the transmission from members to the CH, and tCHtoBSis the time required for the transmission from CH tothe base station. The expected consumed energy of a node to be a CH after a steady-state phase could be represented as (6). (6) All the sensor nodes are assumed to transmit and receive the same size of messages, i.e. l bits of data. The distance to thebase station, dtoBS, could be computed based on the received signal strength. Then, the expected residual energy of a node to be a CH after a steady-state phase could be obtained via,



(7) where the term Eresidual means the residual energy of a sensor node before the process of cluster head selection. Network Traffic is calculated from the average bandwidth usage of all sensor node at the time of CH selection. 3. Proposed Clustering Approach System Assumptions This paper considers network applications in which sensor nodes are deployed randomly in order to continuously supervise the circumstance. The information gathered by sensor nodes is sent to a base station located outside of the deployment field. Each sensor nodes can operate either in sensing mode to monitor the environment parameters and transmit it to the associated CH or in CH mode to collect the data, compress it and resend it to the base station. In addition, some assumptions are made as follows: 1) All sensor nodes and the base station are stationary after deployment. 2) The network is considered homogeneous and all sensor nodes have the same initial energy. 3) Nodes have the capability of controlling the transmission power according to the distance of receiving nodes. 4) The distance between nodes can be computed based on the received signal strength. 5) The radio link is symmetric in such a way that energy consumption of data transmission from node A to node B is the same as that of transmission from node B to node. Handing Uncertainties Using Fuzzy Inference Systems TABLE I: FUZZY MAPPING RULE Residual Network ERE Chance Energy Traffic High High Very Low Very high High Medium Low High High Low Rather Low Medium Rather high High Medium High Rather high Medium Medium Rather high Rather high Low Rather Low Medium Medium High Medium Medium Medium Medium Low Rather High Medium Low High Rather low Rather low High High Very low Rather low Medium Medium Medium Rather low Low High Rather low Low High Rather High Very low Low Low Medium Low Low Medium High Very Low Very low High High Very low Very low Medium Rather High Very low Very low Low Rather High Very low To handle uncertainties, this paper has used fuzzy inference systems (FIS) for the chance computation of each node. As show in Fig. 2, two input attributes for the FIS are the residual energy Eresidual and the expected residual energy EexpResidual, and one output parameter is the probability of a node to be selected as a CH, named chance. The bigger chance means that the node has more chance to be a CH. The

fuzzy set that describes the residual energy input attribute is represented in Fig. 3(a). The linguistic attributes for this fuzzy set are high, rather high, medium, rather low, low, and very low. A trapezoidal membership function is used for high and very low, while a triangular membership function is used for the rest linguistic attributes. Another fuzzy input attribute is the expected residual energy of the CH candidate. The fuzzy set that defines expected residual energy input attribute is illustrated in Fig. 3(b). The linguistic attributes of this fuzzy set are high, medium and rather low, low, and very low. A trapezoidal membership function is used for high and very low, while a triangular membership function is used for the rest linguistic attributes. The other fuzzy input attribute is the expected residual energy of the CH candidate. A trapezoidal membership function is used for the attributes high and low, while a triangular membership function is used for the attribute medium. The only fuzzy output attribute is the chance of a CH candidate. The fuzzy set for the chance output attribute is demonstrated in Fig. 4. Seven linguistic attributes are very high, high, rather high, medium, rather low, low, and very low. The very high and very low have a trapezoidal membership function, and the remaining linguistic attributes are represented by using triangular membership functions. In this work, for simplicity and reducing the cost of computation, the triangular membership functions are mostly chosen here. The chance calculation is accomplished by using predefined fuzzy if-then mapping rules to manage the uncertainty. Based on the two fuzzy input attributes, 18 fuzzy mapping rules are defined in Table I. From the fuzzy rules, we can get the fuzzy attribute chance. This fuzzy attribute has to be changed to a single crisp number that is a form we can use in practice. In our approach, the center of area (COA) method is used for defuzzification of the chance. Generally, fuzzy rules can be generated either from heuristics or from experimental data. In this paper, the heuristic fuzzy rule generation method is used with the principle: A node which holds more residual energy and more ERE has a higher probability to become a CH. Proposed LEACH-ERE Clustering Algorithm Similar to the LEACH, our proposed clustering method configures clusters in every round. The pseudo code of the clustering method is described as Algorithm 1. In everyclustering round (lines 4-30), random number between 0 and 1 is generated by each sensor node. If the random number corresponds to a particular node is bigger than a predefined threshold, which is defined as the percentage of the desired tentative CHs, the node becomes a CH candidate. Then, the node calculates the chance using the fuzzy inference system which is mentioned above and broadcasts a CandidateMessage with the chance.This message means the sensor node is a candidate for CH with the value of chance. Once a node advertises a Candidate-Message, the node waits CandidateMessages from other nodes. If the chance of itself is bigger than every chance values from other nodes, the sensor node broadcasts a CH-Message which means that the sensor node itself iselected as the CH. If a node which is not a CH receives



the CH-Message, the node selects the nearby cluster head as

its CH and sends a JOIN-REQ request to the head.

Fig. 2: Probability reasoning during cluster head selectionof the proposed method Algorithm 1 Proposed Clustering Algorithm Input: N: a network a : a node of N V: {v | v is a’s vicinity node which is a CH candidate} T: a threshold value to become a CH candidatechance(a): a suitability value of the node a to be a CH k: the number of clusters r: the number of times to be a CH Output: CH(a): the cluster head of the node aisClusterHead(a): true if CH(a)=a

Table II Configuration Parameters Type

Network topology

Radio model

Parameter Number of nodes Expected number of clusters Network coverage Base station location Startup energy Eelec ε

Function: broadcast(data, distance); send(data, destination); fuzzylogic(Eresidual, EexpResidual,, network traffic); Initialization: 1. chance(a) ← fuzzylogic( Eresidual, EexpResidual, , network traffic); 2. isClusterHead(a) = false; 3. r ← 0 Main: 4. /* for every clustering round */ 5. if (r== Size[N]/k) 6. isClusterHead(a) ← false; 7. T ← 1; 8. else T← k/Size[N]; 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

end if if (rand(0,1) >T) CH(a) ← a ; chance(a) ← fuzzylogic( EResidual, EexpResidual, network traffic); broadcast(chance(a), V); //Candidate-Message On receiving Candidate-Messages from CH candidates; for each v V if (chance(a) < chance(v) ) CH(a) ← v ; isClusterHead(a) ← false; broadcast(Quit-Election-Message, V) else isClusterHead(a) ← true; r r+1; end if end if if (isClusterHead(a) == true) broadcast(CH-Message, V) On receiving JOIN-REQ messages; else On receiving CH-Message; Send JOIN-REQ messages to the closest CH; end if

4. Performance Evaluation In this section, we present the results of experimental simulations to evaluate our proposed approach. Moreover, we compare our proposed clustering algorithm LEACH-ERE with three different algorithms, namely LEACH [8], LEACHCentralized [8], and CHEF [10]. Simulation results have shown that our approach reveals better performances compared with others.

/ Eelec

Rx

fs

ε

Application

Tx

mp

Simulation times Packet header size Broadcast packet size Data packet size Competition radius Bandwidth

Value 100 5 (0, 0) (100, 100) m at (50, 175) m 2J 50 nJ/bit2 10 pJ/bit/m 0.0013 pJ/bit/m

4

15 25 Bytes 16 Bytes 500 bytes 25 m 1 Mb/s

Simulation Environments The simulation was implemented based on the network simulator, NS-2 [17]. The 100 number of nodes are randomly distributed in a 100 × 100 area. The base station is located at a point (50, 175). The values used in the first order radio model are described in Table II. Simulation Results Handy et al. [18] proposed the metric Half of the NodesAlive (HNA) which denotes an estimated value for the roundin which half of the senor nodes will die. This metric is beneficial in heavily deployed sensor networks. As depicted in Fig. 5, the proposed LEACH-ERE methodology outperforms LEACH and CHEF. LEACH-ERE is more efficient than LEACH about 42.61% and CHEF about 2.87%. LEACH performance is the poorest one, since it does not consider the residual energy level of sensor nodes during clustering. Moreover, the distributed LEACH-ERE has the similar performance as compared with the centralized LEACH-C. Fig. 3 illustrates the distribution of the alive sensor nodes with respect to the number of rounds for each algorithm. This figure clearly displays that our proposed approach is more stable than the other distributed clustering algorithms (LEACH and CHEF), because sensor node deaths begin later in LEACH-ERE and continue linearly until all sensor nodes die. When compared with the centralized clustering approach LEACH-C, the proposed approach has an approximated result without requiring global network knowledge. The following figure shows that LEACH-ERE extend the lifetime of sensor networks. Fig. 6 shows the distribution of the number of clusters with respect to the number of rounds for each algorithm. LEACH-C generates a constant number of clusters until around the round 560 while the numbers of clusters in



LEACH, CHEF, and LEACH-ERE are getting varied. Table III illustrates the average and standard deviation of the number

clusters in LEACH-ERE issteadier than that in other distributed clustering algorithms (LEACH and CHEF). LEACH uses a fully random approach to produce cluster heads, thus it results in a fairly variable number of clusters, even though the anticipated number of cluster heads per round is deterministic.Fig. 8 illustrates the average number of received packets per second at the base station. Obviously, the centralized LEACH-C has the best performance since the base station receives the most information from the sensor nodes during the network lifetime. On the other hand, the proposed distributed LEACH-ERE has the better result as compared with the LEACH and CHEF. 100

80

LEACH-c

60

CHEF

LEACH-ERE 40

Fig. 3: No. of Alive Nodes Vs No. of Rounds 20

0 round 1

round 2

round 3

round 4

Fig. 6: Comparison of LEACH-C, CHEF and LEACH-ERE Average and standard deviation of the number of clusters up to round 600 Ave. Std. Dev.

LEACH

LEACH-C

CHEF

LEACH-ERE

4.2 2.11

4.7 1.03

4.7 1.6

4.7 1.58

Fig.4:CH Selection during 1st round

Fig. 7:Comparison of LEACH, LEACH-C, CHEF and LEACH-ERE for no.of Alive nodes

Fig. 5:CH Selection during 2nd round of clusters up to round 600. It is apparent that the number of

5. Conclusion This novel system proposed Energy efficient technique in designing WSNs. To achieve the energy efficiency, many clustering algorithms are proposed and LEACH is acting as



representative one. LEACH utilizes the probability model to spread the concentrated energy consumption of the CHs. However, it depends on only a probability model and the energy efficiency is not increased. In this work, a fuzzy logic- based clustering approach based on LEACH architecture with an extension to the energy predication has been proposed for WSNs, namely LEACH-ERE. The main objective of our algorithm is to prolong the lifetime of the WSN by evenly distributing the workload among sensor nodes. To attain this goal, we have mostly concentrated on selecting proper CHs from existent sensor nodes. LEACHERE selects the CHs considering expected residual energy of the sensor nodes. The simulation results express that the proposed LEACH-ERE is more efficient than other distributed clustering algorithms,such as LEACH and CHEF. The proposed LEACH-ERE algorithm is designed for the WSNs that have stationary sensor nodes. As a future work, this method can be extended for handling mobile sensor nodes. Also, a further direction of this work will be to find the optimal fuzzy set and to compare the enhanced approach with other clustering algorithms. 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]

G. Anastasi, M. Conti, M. D. Francesco, and A. Passarella, “Energy conservation in wireless sensor networks: A survey,” Ad Hoc Netw., vol. 7, no. 3, pp. 537–568, May 2009.

[3]

L. Atzori, A. Ierab, and G. Morabito, “The internet of things: A survey,” Comput. Netw., vol. 54, no. 15, pp. 2787–2805, Aug. 2010.

[4]

J. S. Lee, “A Petri net design of command filters for semi-autonomous mobile sensor networks,” IEEE Trans. Ind. Electron., vol. 55, no. 4, pp. 1835–1841, Apr. 2008.

[5]

Z. Guo, M. C. Zhou, and L. Zakrevski, “Optimal tracking interval for predictive tracking in wireless sensor network,” IEEE Commun. Lett.,vol. 9, no. 9, pp. 805–807, 2005.

[6]

Z. Wang, L. Liu, M. C. Zhou, and N. Ansari, “A position-based clustering technique for ad hoc intervehicle communication,” IEEETrans. Syst. Man Cybern. C Appl. Rev., vol. 38, no. 2, pp. 201–208, 2008.

[7]

W. R. Heinzelman, A. Chandrakasan, and H. Balakrishnan, “Energy-efficient communication protocol for wireless microsensor networks,” in Proc. IEEE Annu. Hawaii Int. Conf. Syst. Sci., pp. 3005– 3014, Jan. 2000.

[8]

W. B. Heinzelman, A. P. Chandrakasan, and H. Balakrishnan, “An application-specific protocol architecture for wireless microsensor net-works,” IEEE Trans. Wireless Commun., vol. 1, no. 4, pp. 660–670, 2002.

[9]

I. Gupta, D. Riordan, and S. Sampalli, “Cluster-head

election using fuzzy logic for wireless sensor networks,” in Proc. Annu. Conf. Commun. Netw.Services Res., pp. 255–260, 2005. [10] J. Kim, S. Park, Y. Han, and T. Chung, “CHEF: Cluster head election mechanism using fuzzy logic in wireless sensor networks,” in Proc. Int.Conf. Adv. Commun. Technol., pp. 654–659, Feb. 2008. [11] J. Anno, L. Barolli, A. Durresi, F. Xhafa, and A. Koyama, “Performance evaluation of two fuzzy-based cluster head selection systems for wireless sensor networks,” Mobile Inform. Syst., vol. 4, no. 4, pp. 297– 312, 2008. [12] H. Bagci and A. Yazici, “An energy aware fuzzy unequal clustering algorithm for wireless sensor networks,” in Proc. IEEE Int. Conf. FuzzySyst., pp. 1–8, Jul. 2010. [13] H. Ando, L. Barolli, A. Durresi, F. Xhafa, and A. Koyama, “An intelligent fuzzy-based cluster head selection system for wireless sensor networks and its performance evaluation,” in Proc. Int. Conf. Netw.Based Inform. Syst., pp. 55–61, Sep. 2010. [14] G. Ran, H. Zhang, and S. Gong, “Improving on LEACH protocol of wireless sensor networks using fuzzy logic,” Journal of Inf. Comput. Sci., vol. 7, no. 3, pp. 767–775, 2010. [15] Maurya S, Daniel A, “An Energy Efficient Routing Protocol under Distance, Energy and Load Parameter for Heterogeneous Wireless Sensor Networks”, International Conference on Information Technology (ICIT), pp.161 - 166, 2014. [16] Mohan C, Suman, Kumar A, “ Heterogeneous fuzzy based clustering protocol” IEEE international conference on Image Information Processing (ICIIP), pp.601-606, 2013.
