Scholarly article on topic 'LEATCH: Low Energy Adaptive Tier Clustering Hierarchy'

LEATCH: Low Energy Adaptive Tier Clustering Hierarchy Academic research paper on "Computer and information sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Computer Science
OECD Field of science
Keywords
{"wireless sensor networks" / routing / LEACH / clustering / latency / coverage / "energy efficiency" / scalability}

Abstract of research paper on Computer and information sciences, author of scientific article — Wafa Akkari, Badia Bouhdid, Abdelfettah Belghith

Abstract In Wireless Sensor Networks, low latency, energy efficiency, and coverage problems are considered as three key issues in designing routing protocols. In this paper we present a new protocol called Low Energy Adaptive Tier Clustering Hierarchy (LEATCH), which offers a good compromise between delay and energy consumption and resolves some coverage problems. For our purpose, a two level hierarchical approach has been proposed to organize a sensor network into a set of clusters, every cluster divided into small clusters that are called Mini Clusters. As the way the clusters are organized, for each mini cluster we define a Mini Cluster- Head (MCH). Every MCH communicates with the cluster-head directly, it aggregates its mini-cluster information. In addition, we have made some changes in the procedure of cluster head and mini cluster head election. LEATCH promises better performances than the conventional LEACH protocol which is one of the most known hierarchical routing protocols using the probabilistic model to manage the energy consumption in WSNs. Simulation results show that LEATCH performs better than LEACH in term of energy, delay, coverage and scalability.

Academic research paper on topic "LEATCH: Low Energy Adaptive Tier Clustering Hierarchy"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedia Computer Science 52 (2015) 365 - 372

6th International Conference on Ambient Systems, Networks and Technologies,

(ANT 2015)

LEATCH: Low Energy Adaptive Tier Clustering Hierarchy

Wafa Akkaria, Badia Bouhdida, Abdelfettah Belghithb*

aHANA Laboratory, University ofManouba, Tunisia b College of Computer and Information Sciences, King Saud University, Saudi Arabia

Abstract

In Wireless Sensor Networks, low latency, energy efficiency, and coverage problems are considered as three key issues in designing routing protocols. In this paper we present a new protocol called Low Energy Adaptive Tier Clustering Hierarchy (LEATCH), which offers a good compromise between delay and energy consumption and resolves some coverage problems. For our purpose, a two level hierarchical approach has been proposed to organize a sensor network into a set of clusters, every cluster divided into small clusters that are called Mini Clusters. As the way the clusters are organized, for each mini cluster we define a Mini Cluster-Head (MCH). Every MCH communicates with the cluster-head directly, it aggregates its mini-cluster information. In addition, we have made some changes in the procedure of cluster head and mini cluster head election. LEATCH promises better performances than the conventional LEACH protocol which is one of the most known hierarchical routing protocols using the probabilistic model to manage the energy consumption in WSNs. Simulation results show that LEATCH performs better than LEACH in term of energy, delay, coverage and scalability.

© 2015TheAuthors. PublishedbyElsevierB.V.This is an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the Conference Program Chairs

Keywords: wireless sensor networks; routing; LEACH; clustering; latency; coverage; energy efficiency; scalability

1. Introduction

Wireless Sensor networks (WSNs) have received significant attention of researchers in recent years sue to its wide range of applications such as military surveillance, environmental monitoring, forest fire detection, health care and other area1,2. A WSN consists of spatially distributed sensor nodes, which are interconnected without the use of any wires3. In a WSN, sensor nodes sense the environment and use their communication components in order to transmit the sensed data over wireless channels to other nodes and to a designated sink point, referred to as the Base Station (BS). BS collects the data transmitted to it in order to act either as a supervisory control processor or as an access point for a human interface or even as a gateway to other networks4. However, ensuring the direct communication between a sensor and the BS may force nodes to emit their message with such a high power that their resources could be quickly depleted. Therefore, the collaboration of nodes ensures that distant nodes communicate with the BS. In

* Corresponding author. Tel.: +0-000-000-0000 ; fax: +0-000-000-0000. E-mail address: abelghith@ksu.edu.sa

1877-0509 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the Conference Program Chairs

doi:10.1016/j.procs.2015.05.110

this way, messages are propagated by intermediate nodes so that a route with multiple links or hops to the BS is established.

Taking into account the reduced capabilities of sensors, the communication with the BS could be initially conceived without a routing protocol. With this premise, the flooding algorithm stands out as the simplest solution. In this algorithm, the transmitter broadcasts the data which are consecutively retransmitted in order to make them arrive at the intended destination. However its simplicity brings about significant drawbacks. Firstly, an implosion is detected because nodes redundantly receive multiple copies of same data messages containing similar information5.

Moreover, nodes do not take into account their limited power resources when making functional decisions which have great influence on the life time and performance of the network. WSNs distinguish from other wireless networks like mobile ad hoc networks or cellular networks, since WSNs are formed by a significant number of nodes, the manual assignation of unique identifiers becomes infeasible6.

Considering all these functioning properties od WSNs, routing protocols become necessary in WSNs. There are plenty of approaches concerning routing algorithms in wireless sensor networks and they can be divided into flat routing and hierarchical routing in the network structure. All sensor nodes in the flat routing protocol generally have the same function. However, the nodes in the hierarchical routing usually play different roles. The high energy node in some hierarchical routing protocols is used to process and send messages. However, nodes having low energy level are used to sense the target area information. Hierarchical routing protocols proved to be scalable7,20,21. The Common hierarchical routing protocols are: LEACH8, PEGASIS9, TEEN10We pay more attention to LEACH (Low-Energy Adaptive Clustering Hierarchy) protocol, proposed by Wendi B. Heinzelman8.

In this paper, we present hierarchical routing protocol for WSNs, LEATCH (Low Energy Adaptive Tier Clustering Hierarchy), which enhances the LEACH protocol, solves some coverage problems from which LEACH suffers, and finally thrives to offer better performances than LEACH in terms of throughput, delay and power consumption. Thus, we put forward the improvement LEACH mechanism in terms of energy consumption, network coverage, scalability and delay. LEATCH considers two clustering levels for wireless sensor networks in order to guarantee better communication between the BS and the majority of the network nodes.

The rest of the paper is organized as follows: first we give an overview of challenge and design issues in WSNs. In section 3, we highlight the functional inefficiencies inherent to LEACH. We devote section 4 to describe our approach, the LEATCH. In section 5, we present the performance evaluation of our proposed mechanism and its comparison with LEACH. Section 6 provides some concluding remarks and future directions.

2. Challenge and design issues in WSNs

Despite the innumerable applications of WSNs, these networks have several restrictions. The design of routing protocols in WSNs is influenced by many challenging factors. These factors should be overcome so that efficient communication can be achieved in WSNs. In the following, we summarize some of the routing challenges and design issues that affect routing process in WSNs based on some previous studies presented in11,12,13.

Limited energy capacity: Senor nodes are endowed by a limited battery power and random deployment in difficult terrain make it almost impossible to recharge or to replace the dead battery.

Data Aggregation: Since sensor nodes may generate significant redundant data, similar packets from multiple nodes can be aggregated so that the number of transmissions is reduced.

Scalability:Routing protocols should be able to scale with the network size. Also, sensors may not necessarily have the same capabilities in terms of energy, processing, sensing, and particularly communication.

Network Dynamic:The topology of a WSN changes frequently due to sensor addition, deletion, node failures, damages, or energy depletion. Also, the sensor nodes are linked by a wireless medium, which is noisy, error prone, and time varying. Therefore, routing paths should consider network topology dynamics.

Delay: Some applications require that a message must be delivered within a specified time, otherwise the message becomes useless or its information content decreases after the time bound. Therefore, one of the main goals of these protocols is to completely control the network delay.

Limited hardware resources: Sensor nodes have also limited processing and storage capacities, and thus can only perform limited computational functionalities. These hardware constraints present many challenges in software development and network protocol design for sensor networks.

3. LEACH Protocol

Low Energy Adaptive Clustering Hierarchy (LEACH): LEACH is one of the most popular clustering algorithms for WSN14. It forms clusters based on the received signal strength and uses the Cluster Head (CH) nodes as gateways to the BS. All the data processing such as data fusion and aggregation are locally performed within the cluster. LEACH forms clusters by using a distributed algorithm, where nodes make autonomous decisions without any centralized control. Initially a node decides to be a CH with a probability p and broadcasts its decision. Each non-CH node determines its cluster by choosing the CH that can be reached using the least communication energy. The role of being a CH is rotated periodically among the nodes of the cluster in order to balance the load. The rotation is performed by getting each node i to choose a random number T (i) between 0 and 1. A node i becomes a CH for the current rotation round if the number T (i) is less than the following threshold8:

T(i) = p ! ,i e G (1)

1 — (r.mod p)

Where p is the desired percentage of CH nodes in the sensor population, r is the current round number, and G is the set of nodes that have not been CHs in the last 1/p rounds.

LEACH forms one-hopcluster topology where each node can transmit directly to the CH and thereafter to the BS, as shown in Fig. 1.a.

Fig. 1. (a) LEACH Clustering model;(b) Time line Operation of LEACH

3.1. LEACH Algorithm

LEACH protocol provides the conception of round. LEACH runs with many rounds. Each round contains two states: set up state and steady state. In cluster setup state, it forms clusters in self-adaptive mode. However, in steady state it performs data transmission. The time devoted to the second state is usually longer than the time reserved to the first state for saving the protocol payload. Fig. 1.b shows the time line operation consisting of set-up and steady state phases of the LEACH protocol.

3.1.1. Set-up phase.

The CH nodes are chosen randomly among the network nodes during the set up phase and several clusters are formed dynamically15. Initially, each node generates a random number in the range from 0 to 1. If it is less than a threshold, T (i), that node considers itself as a CH for the current round. This decision also involves the past history of the node being CH16.

Once the CH nodes are elected, they broadcast advertisement messages. Based on the received signal strength, each non-CH node chooses its corresponds CH node. Each non-CH node transmits a join request message containing its ID back to its chosen CH node using CSMA. After the set-up phase each CH knows its members.

3.1.2. Steady phase.

Once the clusters are formed, each CH allocates its TDMA schedule to its member nodes. Based on the schedule each member node transmits the sensed data to its correspondent CH node. Once CHs collect all the data from their

members, they transmit the aggregated data8along with their own data to the BS. At the beginning of each round, new CH nodes are elected to form new clusters. Thus the lifetime of the network can be estimated based on the number of rounds.

3.2. Limitations of LEACH

Although LEACH saves nodes' energy and prolongs network lifespan, it still have several shortcomings:

• LEACH is suitable for small size networks because it assumes that all nodes can communicate with each others and are able to reach sink, which is not always true for large size network1.

• At the beginning of each round, LEACH randomly selects the cluster heads without considering the residual energy of these nodes. As a result, the elected CH nodes can have the least energy level and consequently die

very soon17.

• Each cluster head using LEACH directly communicates with the BS no matter the distance is near or not. When the network is huge, the communication between cluster heads and the BS consumes much energy for the long distance transmission. So the lifespan of WSN would be shortened18.

• LEACH provides time slots for each node in the network to transmit data to CHs even though some nodes might not have data to transmit18,19. On the other hand, in LEACH, the size of clusters can increase if the number of cluster heads is reduced wish induces excessive delays introduced by the number of nodes in the same cluster.

• The cluster heads can concentrate on one place and therefore isolated nodes (without cluster head) may be found.

• In LEACH there is no mechanism to ensure that the elected CHs will be uniformly distributed over the network. So all cluster-heads might be concentrate only in one part of the network. In addition, the sizes of clusters can be very different and consequently we can find clusters without any members.

4. Proposed protocol LEATCH

4.1. Basic Idea

This work proposes a two levels hierarchical approach. The first level corresponds on the division of the network into super clusters like the way LEACH does. We just changed the metric used for CH selection. Upon the creation of super clusters, we begin the construction of mini clusters by selecting some mini-cluster heads (MCH) in each super-cluster. The choice of this list of MCHs is based on two criteria; density and the residual energy of the candidate node. By this way the super clusters are divided into small clusters called mini cluster, and each mini cluster is controlled by MCH.

The role of the MCH is to collect data from the members of his mini cluster and to retransmit this data to the SCH. The SCH sends all the received data to the sink. In fact SCHs are the gateways between MCH and the sink. The idea behind this approach of clustering is to overcome the assumption of LEACH, which supposes that any node in the network can communicate directly with the sink, and this is impossible especially in networks deployed in large areas. LEATCH manipulates a two hops inter cluster communication. In fact MCH does not communicate directly with the sink, but they use an intermediate node, the SCH, that is located in an area covered by the base station. In Fig. 2 the cluster based idea presented by LEATCH is given in a more clear view.

4.2. Proposed Algorithm

The LEATCH protocol consists of two phases: the first is the set-up phase, and the second one is the steady-state as illustrated in Fig. 3:

Fig. 2. Cluster-based Hierarchical Model in LEATCH.

Set up phase SÏe pi |step2 ^ Steady phase

^ frame

Tsl Ts2 Ts3 Ts4 Ts5 Ts6 Ts7 TsS TsS ~slO

Fig. 3. Time line operations of LEATCH.

4.2.1. Set-up Phase

Like LEACH, in LEATCH nodes are periodically selected to serve as Super Cluster Head (SCH) or Mini Cluster Head (MCH). At the beginning of each round, the sink broadcast a message to declare the start of the set-up phase, which is divided in two steps: Super Cluster construction and Mini Cluster construction.

4.2.2. Super Cluster Construction:

When receiving a start message from the sink, nodes send their candidate messages to be SCH to the sink, which receives these candidate messages. After collecting location data and energy levels from all the received candidate messages, the sink calculates a probability Pi relative to each node, then it choose the list of nodes with the higher probability.The value of Pi is set as follows:

With K is the number of SCH fixed by the algorithm, E(initial )is the initial energy of the node i, Ei(t)is its residual energy and d is the distance between the sink and the node i. The use of this probability ensures the selection of the nodes with the highest energy level and the closer to the base station to act as SCH (gateway) between nodes in the cluster and the sink. The sink broadcasts its decision of which nodes are selected as SCH in the networks. Once a node is selected as SCH, it broadcasts an advertisement message to the others nodes pushing them to join a SCH and form the corresponding super cluster. Nodes (non-SCH) determines a super cluster to belong to, by choosing the SCH that be reached using the least communication energy (based on the signal strength of each SCH message). In figure 5 the super cluster formation scheme is given in a more clear view.

4.2.3. Mini Cluster Construction:

The joint message received by the SCH from the nodes members, contain information about this node, its density and its residual energy, this two parameters help the SCH to calculate a probability Qi , and based to the value of Qi the SCH decide which nodes will be selected to be MCH. The value of Qi is set as follows:

Qi = ——( ) • q ■ density (3)

—initial

With Eintial is the start energy, Ei(t) is the residual energy and density is the density of the node. The number of the MCH fixed by the SCH depending on the size of the super cluster. The use of the density as a metric do determine the qualified nodes to be MCHs, serves to gather nodes from the same region into mini cluster, and avoid the problem of empty clusters. The SCH broadcast its decision to the list of the MCH, after receiving this message the elected MCHs broadcast advertisement message, using the non-persistent Carrier Sense Multiple Access Medium Access Control (CSMA MAC), to the other node of the super cluster. Based to the signal strength of the advertisement message a simple node determines their mini cluster by sending a joint message to its MCH.

After receiving all the joint messages from members, the MCH create a tdma table and each member of the mini cluster determines its TDMA slot for data transmission and goes to sleep until it's time to transmit data.

4.2.4. The steady-state

Thesteady-statephase begins when sensed data are sent from simple node to MCH, from MCH to SCH and from SCH to the sink.

5. LEATCH performance evaluation

5.1. Simulation setup

To investigate Performance evaluation of both LEATCH and LEACH we used OMNeT++ discrete event simulation platform.OMNeT++ aims at providing a rich simulation platform, and leaves creating simulation models to independent research groups22. We used the following metrics to evaluate the proposed scheme:

• Energy Consumption: the amount of consumed energy by the network in each round.

• Life time system: The number of nodes alive at the end of the simulation time.

• Waiting-Delay:the average of the waiting time relative to all the sent packets considering a cluster node.

• Delay:the average of waiting-delay over all the cluster nodes.

• Percentage of covered nodes: the number of covered node in each round.

The reference network of our simulations consists of 300 nodes distributed randomly across a plain area 200x200 meters. The base station is located at position (0,0), provided with sufficient energy resources. Each node is equipped with an energy source wish is set to 0.3J at the beginning of the simulation. We have set the percentage of MCH and SCH both to 0.05.

5.2. Simulation results

In both LEACH and LEATCH, The data transmission in each cluster is organized by the use the TDMA table delivered by the CH node to the members of its own cluster. In such TDMA table a unit time (eg 0.2 seconds) is affected to each node to transmit its data. Assume that in the network, there are n clusters with Li nodes in each cluster (i [1,n]). So each node needs to wait average (L-1) unit time to transmit its message to the CH . The additional 1 unit time need for cluster head routing to sink. So, in total Avr(L-1)+1 unit time is needed in LEACH. Since the sizes of clusters are different in each round, delay value is also changing. In LEATCH, delay computation is Avr(L-1)+2 unit time, it is the time needed in cluster routing with the addition of the time need for MCH to routing packet to SCH, and the time need by the SCH to retransmit packet to the sink. In addition, the nodes number in each mini cluster is very close due the balanced clustering which makes (L-1) is less than that in LEACH.

(c) Number of nodes alive in LEACH and LEATCH № Percentage of covored nodes in LEACH and LEATCH for 30 rounds

Fig. 4. Performance Evaluation of LEACH and LEATCH.

In LEACH, the number of CH nodes is randomly picked wish makes the number of nodes in the formed clusters obviously unequal. Since the network delay depends on the nodes number in the cluster, the bigger is the cluster size (in terms of nodes number), the more important is the delay. This can lead to significant delays when using LEACH due to the fact that the formed clusters, having unbalanced sizes, can be very dense with many members and consequently many traffic flows which induce significant delays.

However, when using LEATCH, the formed super clusters and mini clusters are more balanced in terms of number of member nodes. We can see, Fig. 4.a exhibits the superiority of LEATCH in attaining less Delay than LEACH.

Fig. 4.b represents the total remaining energy of the network in each round. In addition to its superiority in terms of delay over LEACH, we want to show how power conservative is LEATCH. Figure Fig. 4.b presents the power consumption as a function of the number of rounds. We clearly notice that LEATCH outperforms LEACH in termes of energy during all the studied rounds. Compared to LEACH, LEATCH stands out even when the number of rounds is important. LEACH fails to perform well when we increase the number of rounds which enlarge the difference between the studied mechanisms in terms of enregy consumption. This energy consumption gain is mainly achieved by the two hierarchical routing structures in LEATCH instead of direct transmission strategy relative to LEACH. In addition the use of residual energy when electing the cluster head nodes increase the lifetime system.

Fig. 4.c shows the total numbers of nodes that remain alive over the simulation time. The simulation results show that LEATCH performs better than LEACH. Using LEACH, the first node's death occurs after 951 rounds and near to 1253 rounds, all the nodes are dead. While activating LEATCH, the first node dies after 1250 rounds and all the nodes' energy expire after 1500 rounds.

Finally, we turn to the comparative study of LEACH and LEATCH as a function of isolated nodes. We recall here that an isolated node, for a given round, is a particular node not belonging to any cluster (relatively mini cluster) which makes its unable to send any data during the correspondent round. A covered node, however is not an isolated node which means that its covered by some cluster head (and eventually a mini cluster head). Figure Fig. 4.d portrays percentage of covered nodes as a function of the traffic load. It is inferred that the number of isolated nodes in LEATCH is lower than LEACH which is expected since the major difference between the two mechanisms is the use of two level of hierarchical clustering in LEATCH which increases the probability of any node to be covered by a cluster head or a mini cluster head.

6. Conclusion

In this paper, we proposed LEATCH, a routing protocol based on two layer hierarchical clustering scheme. Simulation results show that our protocol outperforms the classic approach LEACH in terms of delay, energy and coverage. This is achieved by using gateway nodes between cluster-head nodes and the sink with balancing clustering technique. Minimizing the total energy of the network while distributing the cluster uniformly has a great impact on system lifetime and performances.

References

1. AkyildizIF, Weilian Su, Sankarasubrmaniam Y, Cayirci E. A Survey on Sensor Network. In: IEEE Communication Magazine;2002. p. 102-114.

2. Chong C Y, Kumar S P. Sensor Networks:Evolution,Opportunities, and Challenges. In:Proc.IEEE;2008. vol. 52. p. 2292-2330.

3. Yick J, Mukherjee B, Ghosal D. Wireless Sensor network survey. In: The International Journal of Computer and Telecommunications Network-ing;2002. p. 102-114.

4. Kandris D, Tsioumas P, Tzes A, Nikoakopoulos G,Vergados D. Power conservation through energy efficient routing in wireless sensor networks. In: The International Journal of Computer and Telecommunications Networking;2009. vol. 9. p. 7320-7342.

5. Villalba LJG, Orozco ALS, Cabrera A T, Abbas CJB. Routing Protocols in Wireless Sensor Networks. In:Sensors;2009. Vol. 9. p. 8399-8421.

6. Zhou H, Mutka M W, Ni L M. Reactive ID Assignment for Sensor Networks. In: Proceedings of IEEE International Conference on Mobile Ad-Hoc and Sensor Systems;2005. p. 567-572.

7. Haitao Z, Shiwei Z Mutka, Wendu B. A Clustering Routing Protocol for Energy Balance of Wireless Sensor Network based on Simulated Annealing and Genetic Algorithm. In: International Journal of Hybrid Information Technology;2014. Vol.7. p. 71-82.

8. Heinzelman w, Chandrakasan A, Balakrishnan H. Energy-efficient Communication Protocol for Wireless Microsensor Networks. In: Proceedings of the 33rd Hawaii International Conference on System Sciences (HICSS-33);2000. Vol. 8. p. 8020-8030.

9. Lindsey S, Raghavendra C S. PEGASIS: Power Efficient Gathering in Sensors Information systems. In: Proceedingd of the IEEE ICC;2001 .Vol. 3.p. 1125-1130.

10. Manjeshwar A, Agrawal D P. TEEN: A protocol for enhanced efficiency in wireless sensor netwroks. In: Proceedings of the 1st International Workshop on Parallel and Distributed Computing Issues in Wireless Networks and Mobile Computing;2001. p. 189.

11. AlKaraki J, Kamal E. Routing Technique in Wireless Sensor Networks: A Survey. In: IEEE Communication Magazine;2004. vol. 11. p. 6-28.

12. Jun Zheng, Abbas Jamalipour. Wireless Sensor Networks: A Networking Perspective. a book published by A John and Sons, Inc, and IEEEE;

13. Kemal Akkaya, Mohamed Younis. A Survey on Routing Protocols for Wireless Sensor Networks. In: Ad hoc Networks;2005 .vol. 3. p. 325-349.

14. Sipon Miah Md, Insoo Koo. Performance Analysis of ILEACH and LEACH Protocols for Wireless Sensor Networks. In:Journal of information and communication convergence engineering;2012 .p. 384-389.

15. Thein M C M, Thein T. An energy efficient cluster-head selection for wireless sensor networks. In:2010 international conference on IEEE;

2010. p. 287-291.

16. Zhang F, Wang Q. A probability optimize clustering routing algorithm for wsns. In:in Electronics Computer Technology (ICECT);2011. vol. 4. p. 348-352.

17. Li-Qing Guo, Yi Xie, Chen-Hui Yang, Zheng-Wei Jing. Improvement on LEACH by Combining Adaptive Cluster Head Election and Two-Hop Transmission. In:Proceedings of the Ninth International Conference on Machine Learning and Cybernetics, Qingdao;2010. p. 1678-1683.

18. Wairagu G Richard. Extending LEACH routing algorithm for Wireless Sensor Network. In: Data Communications Engineering, Makerere University;2009.

19. Nidhi Barta, Anuj Jain, Surender Dhiman. An Optimized Energy Efficient Routing Algorithm For Wireless Sensor Network. In: International Journal of Innovative Technology and Creative Engineering (ISSS: 2045-8711);2009.

20. Imen Jemili, Abdelfatteh Belghith, Mohamed Mosbah. A Synchronous Tiered Based Clustering Algorithm for large-scale Ad hoc Networks. In: Wireless and Mobile Networking, Springer Boston;2008. p. 41-55.

21. Mohamed Amine Abid, Abdelfettah Belghith. A Synchronous Tiered Based Clustering Algorithm for large-scale Ad hoc Networks. In: Personal and Ubiquitous Computing, Springer-Verlag;2011. Vol. 15. No. 8. p. 799-810.

22. Andrs Varga, Rudolf Hornig. An overview of the omnet++ Simulation environment. In: SIMUTools (ISBN:978- 963-9799-20-2);2008.