I-LEACH: An Efficient Routing Algorithm to Improve Performance & to Reduce Energy Consumption in Wireless Sensor Networks Ahmad Patooghy, Mahdi Fazeli
Zahra Beiranvand
Department of Computer Engineering, Iran University of Science & Technology, Tehran, Iran {patooghy, m_fazeli}@iust.ac.ir
Department of Computer Engineering Borujerd Branch, Islamic Azad University, Borujerd, Iran
[email protected]
Abstract— A large amount of energy in nodes of a Wireless Sensor Network (WSN) is consumed due to the inner-network communications. In this paper, an energy efficient routing algorithm is proposed which saves a significant portion of innernetwork communications energy. To do this, the proposed routing algorithm selects sensor nodes with higher residual energy, more neighbors, and lower distance from the Base Station (BS) as Cluster Head (CH) nodes. Then, it manages sensor nodes appropriately and constructs clusters such a way to maximize WSN lifetime and minimize average energy dissipation per each sensor node. To evaluate the proposed routing algorithm, various simulations have been carried out by using of MATLAB simulator. The proposed routing algorithm is compared to the previous proposed algorithms e.g., LEACH, DBS, and LEACH-C algorithms. Results of the simulations show that the proposed routing algorithm has been improved the WSN performance at least 65%, reduces the energy consumption of the WSN up to 62%, and improves the successfully delivered packet ratio by at least 56% as compared to the previous routing algorithms. Keywords- Wireless Sensor Network; Cluster-based Routing Algorithm; Network Lifetime; Base Station; Cluster Head.
I.
INTRODUCTION
Wireless Sensor Networks (WSNs) are among the widely used types of Adhoc wireless networks [1]. The main objective of WSNs is identifying, receiving, and processing the information within a monitoring area [ 2 ][ 3 ]. WSNs are composed of hundreds or thousands of sensor nodes that are randomly distributed in the monitoring geographical area [4][5].It dose not require a fixed network support, so it can be widely used in many fields such as military reconnaissance, medical care, monitoring the physical or environmental phenomenon like: temperature, sound, vibration, pressure, and motion at different locations, and other special areas [6][7]. The four key units seen in architecture of every sensor nodes include sensing unit, processing unit, radio unit, and the power unit [ 8 ]. The sensor nodes are supplied by nonrechargeable batteries mounted on sensors [9][10]; therefore, minimizing energy consumption in order to extend the lifetime of network is of importance issue in WSNs. Since the major portion of energy consumption in sensor nodes is due to radio communications [2], selection of an efficient routing algorithm
significantly reduces the communication energy [1]. Routing algorithms can reduce the energy consumption of WSNs through selection of minimal routes [8][11]. Among the several routing algorithms proposed for WSNs, cluster-based algorithms are more effective in meeting WSNs requirements mainly energy consumption [2][5][9]. By clustering of sensor nodes into some groups called clusters, sensor nodes of each cluster send their data merely to specific sensor nodes in the cluster called Cluster Heads (CH). Then, CH nodes transmit gathered information to the Base Station (BS) [7][12]. Since CH nodes play an important role in the performance of clusterbased routing algorithms [13], the policy of CH node selection deeply affects network parameters i.e., network lifetime, energy consumption rate, and packet delivery delay [6][14]. Several cluster-based routing algorithms are proposed in the literature that among them, LEACH1 algorithm [15] is the most famous algorithm. This algorithm uses a random model to selects CH nodes then, non-CH nodes join to the clusters using one-hop transmissions with Time-Division Multiple Access (TDMA). LEACH algorithm does not consider the remaining energy and geographical position of sensor nodes in the CH selection process [2][6]. This leads to the early death of sensor nodes and the decrease of WSN lifetime. PEGASIS algorithm [ 16 ] forms chains of sensor nodes rather than clusters to transfer information packets to the BS, so that each sensor node sends receives data from its close neighbor. However, high complexity of this algorithm discourages designers to use it in real applications. HEED algorithm [14] extends the basic idea of LEACH algorithm by incorporating residual energy and sensor node proximity to its neighbors in the CH node selection; but it does not pay any attention to the distribution or density of sensor nodes. LEACH-C algorithm [4] uses same steady-state algorithm as LEACH algorithm, but it organized clusters centralized by BS, unlike LEACH algorithm; this can indeed select a reasonable CH node, but it will increase energy consumption greatly as a result of exchange of data between the BS and the sensor nodes. In DBS algorithm [17], adaptive clustering of sensor nodes is based on their distance from the BS. In this algorithm,
1
Low-Energy Adaptive Clustering Hierarchy
sensor nodes death rate is high, and all sensor nodes die in a short time after the death of the first sensor node. In this paper, a cluster-based routing algorithm is proposed that selects CH nodes based on sensor nodes factors geographical position, energy remaining, and number of neighbors. It also manages the sensor nodes and clusters to reduce the energy consumption within the WSN. Similarly, in the proposed algorithm, the operation of selecting the route with minimal energy cost is effective in increasing the WSN lifetime. The results of the simulations show that the proposed algorithm has been improved the network lifetime, reduces the energy consumption of the network, and improves the successfully delivered packet ratio as compared to the previous algorithms. The rest of the paper is organized as follows. In Section II of the paper, a number of the most important cluster-based routing algorithms are discussed. Then, in Section III, the proposed routing algorithm is described. The performed simulations and their results are presented in section IV, and finally, the section V provides a conclusion of the paper and for further directions. II.
RELATED WORK
Network topology and routing algorithm have a great impact on the performance of WSNs. Recent researches carried out in the field of cluster-based routing algorithms, have investigated the algorithms from different perspectives. Among the previously proposed algorithms, the LEACH algorithm [15] has significantly increased the WSN lifetime in comparison with previous non-clustered routing algorithms. It uses the CH nodes as routers to the BS and all data processing e.g., data fusion and aggregation are done locally to the clusters. In each round of LEACH algorithm, the initialization and stabilization phases are performed. During the initialization phase, CH nodes are selected among sensor nodes; after that, other nodes are joined to their nearest CH node in terms of distance and thereby clusters are formed. In the process of CH nodes selection, each sensor node will generate a random number between 0 and 1. If the random number generated is less than the T(n) threshold, where T(n)= / 1 1/ , then the node is selected as CH node. In this equation, p is the probability of selecting nodes as a CH node in the sensor populations; r is the current round number; and (r mod (1/p)) is on behalf of the number of nodes which was elected as CH node in the round r. The sensor nodes within each cluster transmit the received data from the environment to their CH nodes in the slot assigned to them by CH node. In the stability phase, the CH nodes transmit gathered information to the BS. Because of the hierarchical nature of the LEACH algorithm, the operation of the route selection is simple and the routing of data, does not require much information. Furthermore, according to the random mechanism of CH node selection, the opportunities of all sensor nodes to be CH node are equal. However, the LEACH algorithm also has some disadvantages; such as the energy of CH nodes which are away from the BS has been quickly evacuated. Therefore, the development of network under the LEACH algorithm is not feasible. The other is the LEACH algorithm has not consider the sensor node position
factor, so it does not makes the appropriate CH nodes distribution [2]. Energy factor in the LEACH algorithm are not considered in the CH nodes selection. Therefore, the probability of selecting sensor nodes with different energy levels is equal. PEGASIS algorithm [16], an approximate optimal algorithm based on chain structure of sensor nodes that improved LEACH algorithm; in this algorithm the operations of chain implement and chain control can be done by both the BS and the sensor nodes. Sensor nodes on the chain are merely connected with their close neighbors and receive from and transfer to them and sensor nodes take turns transmitting to the BS; thus the average of spent energy by each node per round has reduced. But under the PEGASIS algorithm, all nodes need to have a general knowledge of the network to form chains and greedy algorithms are used. This problem increases PEGASIS algorithm spent cost and makes it difficult to implement of it. HEED algorithm [14], expands LEACH algorithm scheme by applying a combination of factors like remaining energy and number of neighbors. In HEED algorithm, unlike LEACH, CH nodes are well distributed in the network. But, this algorithm does not consider any assumptions about the sensor node capabilities, such as geographical position. LEACH-C algorithm [4], forms better clusters via the dispersing of CH nodes within the network. Besides, during the initialization phase of LEACH-C algorithm, each sensor node sends information about its current position and residual energy level to the BS. Therefore the BS manages the clustering process having a more efficient knowledge of the network. However, a GPS1 is needed to send basic information. DECR algorithm [7] examines the distance between the clusters and the BS by a multi-hop and dynamic method and in this way it solves the problem of the unbalanced distribution of the load. The distance of sensor nodes from the BS has been considered in the policy of clustering mechanism of DBS algorithm [17]. In this way that the sensor nodes with more distance from the BS lessen selected as a CH node than the sensor nodes closer to the BS. Therefore, the sensor nodes have uniform energy consumption compared to the other algorithms. But in areas that sensor nodes are dense, the DBS algorithm haven’t appropriate performance and the overhead of clustering at the beginning of each round of this algorithm is high. III.
PROPOSED ROUTING ALGORITHM
As mentioned in Section II, most of the previous algorithms do not consider information and properties of sensor nodes e.g., energy, location, distance to BS, and the number of neighbor in the selection of CH nodes, cluster formation and data transfer phases. This section, proposes an efficient algorithm which considers these issues to offer a significant reduction in energy consumption. Similar to previous works [2][13][12][18], the proposed routing algorithm is applicable for a WSN which has following specifications: •
1
All nodes have the same initial energy and their batteries are not rechargeable. Sensor nodes die with the end of their battery.
Global Positioning System
•
Sensor nodes are fixed i.e., when they are first randomly distributed, they remain in the same place.
•
Each sensor node has a unique ID and knows its current position and its remaining energy.
•
Each sensor node has enough power to communicate directly with the BS.
•
In a cluster, nodes can obtain various sensory data.
•
Every sensor node has a computational unit and a memory unit. Nodes can run some processes, including the definition of sensed data and aggregated data, etc.
Similar to other cluster-based routing algorithms, three phases of CH nodes selection, cluster formation, and data transfer should be done by the proposed routing algorithm. In the first phase, the CH nodes must be selected among the sensor nodes of WSN. In the second phase, according to the selected CH nodes, clusters should be formed, i.e., non-CH nodes should be appropriately assigned to CH nodes. Finally, in the third phase, data transferring from the sensor nodes to the CH nodes and subsequently from CH nodes to the BS must be done. Details of each phases of the proposed routing algorithm are described in sections III.A to III.C. A. CH Node Selection Phase In order to select the appropriate CH nodes in the CH nodes selection phase, the Improved LEACH (I-LEACH) algorithm takes important factors such as the amount of residual energy of each sensor node, position of the sensor node relative to the BS and the number of neighbors of each sensor node into account. The CH nodes selection directly affects the performance factors of WSN such as load distribution, energy efficiency, and network lifetime [1][14]. The first factor to be examined in the CH nodes selection phase is the number of neighbors of each sensor node. The sensor nodes in the radius of neighborhood of each sensor node are considered as neighbors of that sensor node. The radius of neighborhood is calculated according to (1) RCH=
M
M / π
k
(1)
where, RCH is the radius of neighborhood, M×M is the area where sensor nodes are distributed, and k is the number of clusters [1]. In the CH nodes selection process of I-LEACH algorithm, a sensor node with more neighbors has higher chance to be selected as a CH node. In the I-LEACH algorithm the optimal number of clusters or kopt is considered for the k value. The kopt can be calculated by (2) [1]. kopt =(√N/√2π)( dtoCH = M/(√2k π)
/ and
)(M/
)
(2)
dtoBS = 0.765 (M/2)
In (2), dtoCH is the average distance between sensor nodes and their CH nodes, dtoBS is the average distance between CH nodes and the BS, and N is the number of sensor nodes in the WSN. Most of the previously proposed algorithms do not pay attention to remaining of sensor node energy. Therefore, to improve the selection of CH nodes, the remaining energy of
sensor nodes has also been considered in CH selection. To do this, in each round of the of I-LEACH algorithm, sensor nodes with higher residual energy than the average energy of the all sensor nodes have more chance to be selected as a CH node. Likewise, due to the importance of the selected CH node distance from the BS, the ratio between the average distance of WSN sensor nodes from the BS and the distance between nodes to the BS has also been considered. The more is the node away from the BS, the less is its chance of being selected as CH node. In CH nodes selection phase of I-LEACH algorithm, each sensor node generates a random number between 0 and 1. Then the random number is compared with improved threshold obtained from (3). If the random number is less than the T(n) threshold, then the sensor node is selected as CH node and leaves G set to prevent it from reelection as CH node in the current round. G is a set of sensor nodes that will determine which of the sensor nodes is not selected as a CH node in the last 1/p rounds T(n) =
1
1
(3)
0 where, Ecur is current sensor node power; Eavg is the average energy of the network in the current round; Nbrn is the number of neighbors for n; Nbravg is the average number of neighboring nodes in the network; dtoBSavg is the average distance of the network sensor nodes to the BS; and dtoBSn is distance of sensor nodes from the BS. The more is the distance of the sensor node far from the BS, the more is the amount of energy spent on sending data to the BS. Like node E in Fig.1 (letters represent the nodes and numbers represent CH nodes) if the distance between network sensor nodes from the BS is more than specific degree, selecting sensor nodes as CH nodes at intervals is not cost effective. B. Cluster Formation Phase In most of the previous algorithms, when a sensor node starts the cluster selection phase, the distance of the sensor node from the BS is the only factor examined. Through this selection, the sensor node may join to the CH node that is far from the BS. I-LEACH algorithm, checks the distance from CH node to the BS as well as the distance of sensor nodes from the CH node in the cluster formation phase. Since the distance between the CH node and BS is usually much greater than that of node and CH, much more energy is spent on data transfer in this part. In addition, as seen in Fig. 1, it is possible that a state occurs in which the selected CH node (No. 2) is close to the node (B), but far from the BS; on the other hand, it is possible that another CH node (No. 1) is slightly farther from node (B) but is closer to the BS than the selected CH node (No. 2). In the I-LEACH algorithm, the priority of selection is based on the distance of the CH node from the BS. Thus, the additional energy consumption of remote CH node data transmission to the BS is significantly reduced. Considering these distances increases precision. So a sensor node joins to the CH node for which the output of (4) for the selected CH nodes has the minimal amount.
(6)
ERX(k) = kEelec In (5), Eelec, is the energy consumption of electronic circuits.
Figure 1.
B. Simulation Settings In order to evaluate performance of the I-LEACH algorithm, a MATLAB simulator has been used. Table 1 contains the values of environmental parameters used in the simulation [17]. Simulated network model is shown in Fig. 3. CH nodes are shown with the colored circles and empty circles represent non-CH nodes. The marked × is showing the location of the BS.
An Example of WSN for Implementing I-LEACH Algorithm.
dtoBSC × dtoCH
(4)
In the (4), C is the examined CH node, dtoBS is the distance of the CH node from the BS; and dtoCH is the distance of the S node from the CH node C. The CH node creates a TDMA schedule and assigns each its member node a time slot when it can transmit. During this phase, the sensor nodes can begin sensing and transmitting data to the CH nodes. C. Data Transmition Phase In this phase of the I-LEACH algorithm, the CH nodes collect the obtained data from their members and integrate them. Since those sensor nodes which are close to each other, have high overlapping and receive similar data from the environment, their integration reduces the cost of the extra data transmission greatly. Then, CH nodes send the obtained information to the BS. To avoid the interference of transmitted signals by the CH nodes, CDMA1 , is assigned to every CH node; but, there may be some sensor nodes within the network which are closer to the BS than all CH nodes [19], such as sensor node F in Fig. 1. I-LEACH algorithm is adjusted in such a way that nodes with these conditions send their data to the BS directly. By doing this, the cost to join and become a member of CH nodes which are far from the BS, is dropped. In fact this action will reduce the extra transfers and similarly, use the benefits of planar algorithms inside the hierarchical algorithm for sending data efficiently to the BS. IV.
The Radio Model of Energy Consumption for Sending and Receiving Packets.
EVALUATION OF THE I-LEACH ALGORITHM
A. Used Energy Model Like most of previous papers [4][8][13][16][18], paper the radio model of energy consumption of the network has been consider ed, as shown in Fig. 2. In this model, k represents the transmitted data packet size; d is the transfer distance; ETX is the required energy consumption for data transmition which is dependent on the two parameters of k and d, and ERX is the required energy consumption for data receiving that is dependent on the value of k. The following equations are energy consumption relations. , ,
ETX(k,d) = d0 = 1
Figure 2.
Code Division Multiple Access
/
(5)
[
200 180 160 140 120 100 80 60 40
,
20 0
0
20
40
60
80
100
120
140
160
Figure 3. The Initial Simulated Network.
180
200
200
3000
Rounds
2000
180
DBS
160
LEACH-C
140
I-LEACH 1500
Alive Nodes
2500
LEACH
1000
LEACH DBS LEACH-C
120
I-LEACH
100 80 60 40
500
20 0
0 LND
MND
500
FND
1500
2000
Rounds
Figure 4. Average time of first, mid, ,and last sensor node to die in the proposed algorithm versus other algorithms.
Figure 5. WSN Lifetime in the proposed algorithm versus other algorithms. 200
0,5
180
0,45
LEACH
0,4 0,35 0,3
LEACH-C
0,25
I-LEACH
LEACH
160
DBS
DBS
140
Data Packets
Average of WSN Energy
1000
0,2 0,15
LEACH-C
120
I-LEACH
100 80 60
0,1
40
0,05
20
0 0
500
1000
1500
2000
Rounds
0 500
1000
1500
2000
Rounds
Figure 6. Average of WSN Energy Consumption in the proposed algorithm versus other algorithms.
C. Simulation Results & Analysis In order to evaluation of proposed algorithm, the algorithms with the similar basic idea are selected. The LEACH algorithm as the basis of cluster-based routing algorithm is selected in order to compare with the proposed algorithm. Beside it, DBS and LEACH-C algorithm have been selected. DBS algorithm is one of the algorithms that its CH selection and cluster formation policies are designed based on the distance of sensor nodes from the BS. LEACH-C algorithm is another algorithm that is presented by LEACH algorithm designers and its measure for CH node selection is remaining energy of nodes. The results of simulation are compared with these algorithms and are presented in the chart format. Fig. 4 shows the evaluated parameters i.e., FND 4 , MND 5 , and LND 6 in the simulation. It can be seen in some cases that the time of the first node death in the I-LEACH algorithm is earlier than three other algorithms and thus the network stability duration is partly reduced. In contrast, 62 percent reduction in the energy consumption than the LEACH algorithm has resulted in the significant decrease of the mortality of sensor nodes in the ILEACH algorithm and the network lifetime has increased up to 67%. Furthermore the value of AAN7 parameter for I-LEACH algorithm has been increased about 53% than LEACH algorithm, 45% than DBS algorithm, and 15% than LEACH-C algorithm. Therefore, this issue is negligible with respect to the overall network performance for the applications where stability is not so important. As can be seen in Fig. 5, the important factors such as location, energy and number of neighbors are considered in the 4 5 6 7
First Node Death Mid Node Death Last Node Death Average of Alive Nodes
Figure 7. Number of Delivered Data Packets in the proposed algorithm versus other algorithms.
CH nodes selection, in addition lead to the CH nodes optimal selection, directly affects the improves the efficiency of routing algorithm and extends the WSN lifetime. The network lifetime with I-LEACH algorithm is increased about 67% than LEACH algorithm, 59% than DBS algorithm, and 29% than LEACH-C algorithm. Fig. 6 shows the energy diagram of I-LEACH algorithm compared with other algorithms. Energy distribution and load balancing is done simultaneously, using CH nodes selection management strategies and restrictions of the sensor nodes that have a membership fee in the clusters and limiting the sensor nodes that have the capability of being CH node. In other words, the total energy consumption in the network is optimally managed. The energy consumption in I-LEACH algorithm has been 62% reduced compared with LEACH algorithm, 55% than DBS algorithm, and as well as 27% than LEACH-C algorithm. As well as the number of data packets that are delivered to the BS has been increased up to 56%using I-LEACH algorithm compare to the LEACH algorithm, 46% than DBS algorithm, and 21% than LEACH-C algorithm, Fig. 7. Since in some of the WSN implementations, the BS is placed in a remote point of the sensor-covered area, the simulation is repeated by changing the location factor of BS and placing it in an area outside the WSN. As seen in Fig. 8, the I-LEACH algorithm improves the network lifetime to 55% than LEACH algorithm, 34% than DBS algorithm, and 62% than LEACH-C algorithm by changing the position of the BS from the center of network to 100×175 and as well as reduces the energy consumption to 59% than LEACH algorithm, 40% than DBS algorithm, and 39% than LEACH-C algorithm, Fig. 9. Furthermore, using the I-LEACH algorithm, the stability of the WSN is increase and in this way, it has good effects on the performance of these networks.
200 0,3
180 LEACH
160
Alive Nodes
Average of WSN Energy
DBS
140
LEACH-C
120
I-LEACH
100 80 60 40
LEACH
0,25
DBS 0,2
LEACH-C I-LEACH
0,15 0,1 0,05
20 0
0 100
300
500
700 900 Rounds
1100
1300
1500
Figure 8. The WSN Lifetime in the proposed algorithm versus other algorithms with moving BS outside the Network.
V.
CONCLUSIONS
Since energy consumption plays an important role in WSNs, energy efficiency and load balancing are significant challenges of routing algorithms in WSNs. In this paper, an energy efficient routing algorithm for WSNs was proposed which considers sensor nodes current energy, sensor nodes position, and the number of neighbors of sensor nodes in the selection of CH nodes. By selection of sensor nodes that have more remaining energy, and also have the relatively little distance for being CH node, as well as by tacking into account the distance of CH nodes from the BS factor in the cluster formation phase, 1) the average energy consumption of the WSN is reduced; 2) the WSN lifetime is extended as compared to LEACH, DBS and LEACH-C algorithms. Simulation results supports that the I-LEACH algorithm has the flexibility of changing the position of BS in WSN. REFERENCES [ 1 ] O. Zytoune, D. Aboutajdine, M. Taz, “Energy balanced clustering algorithm for routing in heterogeneous wireless sensor networks”, 5th International Symposium on I/V Communications and Mobile Network (ISVC), pp. 1–4, Sept. 30 2010-Oct. 2 2010. [2] Q. Bian, Y. Zhang, Y. Zhao, “Research on clustering routing algorithms in wireless sensor networks”, IEEE International Conference on Intelligent Computation Technology and Automation (ICICTA), Vol. 2, pp. 11101113, 11-12 May. 2010. [3 ] J. Lin, M. Liao, “A clustering patch hierarchical routing protocol for wireless sensor networks”, The 5th IEEE International Conference on Computer Science & Education Hefei, pp. 941-948, 24–27 Aug. 2010. [4] W. B. Heinzelman, A. P. Chandrakasan, H. Balakrishnan, “An application -specific protocol architecture for wireless sensor networks”, IEEE Trans. On Wireless Communications, Vol. 1(4), pp. 660-670, Oct. 2002. [5] Q. Yang, Y. Zhuang, H. Li, "A multi-hop cluster based routing protocol for wireless sensor networks", Journal of Convergence Information Technology, Vol. 6(3), March, 2011. [ 6 ] L. Shujiang, D. Kuan Kong Lixin, “Improvement and simulation of clustering routing algorithm in wireless sensor network”, 6th IEEE Conference on Industrial Electronics and Applications (ICIEA), pp. 1068–1073, 21-23 June. 2011. [7] B. Lotfimanesh, N. Seyedgogani, “DECR: dynamic and energy effective clustering based routing algorithm for prolonging the lifetime of wireless sensor networks”, IEEE International Conference on Advanced Mechatronic Systems (ICAMechS), pp. 425–428, 11-13 Aug. 2011. [ 8 ] D. Peng, Q. Zhang, “An energy efficient cluster-routing protocol for wireless sensor networks", IEEE International Conference on Computer Design and Applications (ICCDA), Vol. 2, pp. V2-530–V2-533, 25-27 June. 2010.
200
700
1200 Rounds
Figure 9. Average of WSN Energy Consumption in the proposed algorithm versus other algorithms with moving BS outside the Network. [9] M. Haneef, Z. Wenxun, Z. Deng, “MG-LEACH: Multi group based leach, an energy efficient routing algorithm for wireless sensor network”, 14th IEEE International Conference on Advanced Communication Technology (ICACT), pp. 179-183, 19-22 Feb. 2012. [10] H. Duc Chinh, C. Kuan Thye, R. Kumar, A. Kumar Panda. “Intra-cluster power management in hierarchical routing protocols for wireless sensor networks”, IEEE International Symposium on Industrial Electronics, pp. 1147-1152, 28-31 May 2012. [11] Ch. Meng-xiao, H. Ting-Lei, Li. Hai-ming, X. Hong-bo, “Research of routing algorithm for wireless sensor network”, IEEE International Conference on Intelligent Computing and Integrated Systems (ICISS), pp. 890-893, 22-24 Oct. 2010. [12] A. Ahmed Abbasi, M. Younis, “A survey on clustering algorithms for wireless sensor networks”, Elsevier B.V. 2007, Network Coverage and Routing Schemes for Wireless Sensor Networks, Computer Communications, Vol. 30, pp. 2826-2841, 15 Oct. 2007. [13] M.Ch.M. Thein, Th. Thein, “An energy efficient cluster-head selection for wireless sensor networks”, IEEE International Conference on Intelligent Systems, Modelling and Simulation, pp. 287-291, 27-29 Jan. 2010. [14] O. Younis, S. Fahmy, “HEED: A hybrid, energy-efficient, distributed clustering approach for Ad hoc sensor networks”, IEEE Transactions on Mobile Computing, Vol. 3(4), pp. 366-379, Oct.-Dec. 2004. [ 15 ] W. Heinzelman, A. Chandrakasan, and H. Balakrishnan, “Energyefficient routing protocols for wireless microsensor networks”, 33rd Hawaii International Conference System Sciences (HICSS), Jan. 2000. [16] S. Lindesey, C. S. Raghavendra, “PEGASIS: Power-Efficient Gathering in Sensor Information Systems”, Proc. Aerospace IEEE Conference, Vol. 3, pp. 3-1125 – 3-1130, 2002. [17] N. Amini, M. Fazeli, S. G. Miremadi, M. T. Manzuri, “Distance-Based Segmentation: An Energy-Efficient Clustering Hierarchy for Wireless Microsensor Networks”, Proc. 5th Annual Conference on Communication Networks and Services Research, Fredericton, pp. 18-25, May 2007. [18] N. Sun, N-b. Su, S-h. lee, “A lifetime extended routing protocol based on data threshold in wireless sensor networks”, 10th IEEE International Conference on Computer and Information Technology (CIT), pp. 743– 748, June. 29 2010 - Jul. 1 2010. [ 19 ] E. Arbab, V. Aghazarian, A. Hedayati, N. Ghazanfari Motlagh, “A LEACH-based clustring algorithm for optimizing energy consumption in wireless sensor networks”, 2nd IEEE International Conference on Computer Science and Information Technology (ICCSIT), Singapore, 28-29 Apr. 2012.
