Energy Aware Clustering Aggregate Node Rotation with Sink Relocation (EAC-ASR) protocol for MANET

Energy Aware Clustering Aggregate Node Rotation with Sink Relocation (EAC-ASR) protocol for MANET

First A. Author, Designation, Organization, Second B. Author, and Third C. Author, Jr., Designation, Organization

Abstract—Mobile Ad-hoc Networks (MANETs) are currently deployed for various applications, as a result of significant improvement in the technological development, diverse sensing and mobility capabilities. Reduction in the energy consumption is the most important challenge in the MANET, to improve the communication efficiency at the individual nodes. An Energy Aware Clustering Aggregate Node Rotation with Sink Relocation (EAC-ASR) protocol is proposed to enhance the energy efficiency of the MANET. Energy aware clustering process is done to improve the access control mechanism of the network. Aggregation of data from the nodes is performed using the data collection algorithm. Rotation of the mobile nodes and relocation of the sink are performed, to balance the energy consumption in the network, during the data transmission process. The simulation results show that the proposed EAC-ASR protocol reduces the energy consumption and increases the network lifetime.

Index Terms—Data Aggregation, Energy Aware Clustering Aggregate Node Rotation with Sink Relocation (EAC-ASR) protocol, Mobile Ad-hoc Network (MANET), Mobile Node Rotation

U

I. Introduction

sage of the portable and high speed computing and communication devices are increased nowadays, owing to the advancement in the Wireless Communication Technologies. The Mobile Ad-hoc Network (MANET) is suitable for the cost-effective and quick network setup within a short period for strategic communication in the Military and rescue operations. Clustering is a most significant research area in MANET, because it improves the performance of flexibility and scalability when the network size is huge with high mobility. Energy consumption becomes an important issue in the MANET.

This paper proposes an Energy Aware Clustering Aggregate Node Rotation with Sink Relocation (EAC-ASR) protocol that enhance the energy efficiency of the network. Deployment of mobile node is carried out by the Network formation, division of regions, calculating the number of nodes, coverage area and probability calculations for the divided regions. Energy aware clustering process reduces the energy consumption and improves the access control mechanism of the network. Then, the data aggregation is performed by the data collection algorithm that leads to effective multi-hopping process. Rotation of the mobile nodes and relocation of the sink are performed, to balance the energy consumption in the network, during the data transmission process. Hence, continuous working of the particular hop nodes is reduced. Node rotation and sink relocation process is performed, based on the energy level of the nodes. The theoretical and simulation analysis show that the proposed EAC-ASR protocol reduces the energy consumption and increases the network lifetime.

II. Related Work

III. Proposed Method

This section describes about the proposed EAC-ASR protocol for enhancing the energy efficiency of the MANET. In the proposed technique, Deployment of mobile node is carried out by the Network formation, division of regions, calculating the number of nodes, coverage area and probability calculations for the divided regions. Energy aware clustering process is done to improve the access control mechanism of the network. Aggregation of data from the nodes is performed using the data collection algorithm. Rotation of the mobile nodes and relocation of the sink are performed, to balance the energy consumption in the network. The performance of the proposed protocol is evaluated using various performance metrics.

A. Energy Aware clustering process

Clustering ensures scalability and load balancing in the MANETs and increases the system capacity by facilitating the spatial reuse of resources. Also it elects a cluster head for the enhanced coordination of the transmission activities. This reduces the transmission collision of mobile nodes, to ensure the energy saving and reduced resource consumption. Generation and spreading of the routing information are controlled by forming a virtual backbone for inter-cluster routing including cluster heads and cluster gateways. Therefore, each node stores and processes a fraction of the total network routing information, thus saving a lot of resources. Energy aware clustering is adopted to enhance the energy efficiency of the network. Generally, the mobile nodes are deployed randomly in a specific region. The distance from a node to its cluster head or sink is less than or equal to d₀. Dissipation of energy in the cluster head in a single round is given by Equation (1),

(1)

Where B is the number of bits in the message, D_AVG is the average distance between base station and cluster head and E_DA is the required energy for data fusion or aggregation in a single round. Energy consumed in the non-cluster head is given by Equation (2),

(2)

Where D_CH represents the average distance of the node from the cluster head. The total amount of energy consumed in the cluster is given by Equation (3),

(3)

The total energy dissipatation level of the network is given by Equation (4),

(4)

The optimal number of the clusters can be calculated by finding the derivative of E_Total with respect to k and equating it to zero.

(5)

(6)

(7)

The optimal probability of a node to become as a cluster head is specified by the Equation (8),

(8)

Election of the cluster heads for normal nodes is performed using a probability scheme, based on the average energy and residual energy of the normal nodes. Let, n be the number of nodes and m be the fraction of the number of the nodes with β times more energy than normal nodes. Powerful nodes are known as advanced nodes, and the rest (1 - m) × n as normal nodes. The initial energy of each normal node is E_init and advanced node has E_init× (1 + β). Intuitively, advanced nodes have to become CHs more often than normal nodes, since the energy of the advanced nodes is greater than the energy of the normal nodes. The value of P_opt does not change, but the total energy of the network is changed. The total initial energy of the heterogenous network is given by (8),

(9)

E_R and E_NAVG denotes the residual energy and average energy of a normal node. Since the threshold calculation depends upon the average energy of normal sensor nodes in a round r, therefore it should be calculated. The average energy of normal nodes is estimated as:

(10)

Here R represents the total round of the network lifetime and R can be estimated as

(11)

D.E_round denotes the total energy dissipated in a round of the network and E_N is the total energy of normal nodes in the network. Cluster head threshold for the normal nodes are multiplied by the ratio of residual energy and average energy of the normal nodes in a round, since the energy of the normal nodes is less when compared to advanced nodes. Hence, the normal nodes will become a cluster head, only when it have sufficient energy.

B. Data Aggregation process

The Cluster head (CH) is responsible for the aggregation of data received from the cluster members, and then send the aggregate data to the Base Station (BS) or neighboring CH through the gateway. The cluster node gathers data and send to the CH, if the energy of the CH is below the average energy. The cluster- head is selected based on the energy. The associate cluster head is selected, when the energy of the CH is below the energy of the non-CH nodes. The size of the aggregated packet does not depend on the number of packets aggregated during data fusion, irrespective of the number of nodes in the cluster.

Consider a cluster with a single cluster head node and ‘n’ sensor nodes. The node density is assumed to be constant, hence the number of nodes in each cluster ‘n’ is proportional to the area of the cluster. During each data collection process, the cluster head receives “n” packets from the nodes in its cluster, performs data aggregation and produces (n) packets of the same length. Thus, the number of the output packets is a function of the number of the input packets. Then, the number of packets in the aggregated output is

(n)-cn+h (12)

In this model, h corresponds to the overhead of aggregation, while c is the compression ratio.

1) Cluster head data collection

ClusterHeadDataCollection ()

{

• Association of the number of nodes with various parameter or node parameters.
• Aggregation of all nodes at cluster level.
• Collecting the parameters useful for node information and storing at each cluster head.
• Cost Evaluation on the basis of collected parameters.
• Evaluation of Minimum global cost.
• Sending all cost parameters to the cluster head for further association.
• Transferring the cost parameters to the base station.

}

C. Mobile Node Rotation with Sink Relocation

The mobile nodes are rotated to balance the energy consumption in the network. A node at the high energy consumption location swaps its position with a node at the low energy consumption location. Here, multiple mobile sensors swap their positions once or multiple times. The three nodes initially at locations S₁, S₂ and S₃ consume more energy than the nodes at other locations. The node at S₃ consumes more amount of energy because it is located far from its parent node at S₁. Using the mobile node rotation, multiple nodes are rotated through high energy consumption locations. From the Fig.2, the nodes at the tailback locations S₁, S₂, S₃ can rotate with the nodes at the locations S₈, S₇ and S₅ respectively after a specific time period, to balance the energy consumption between the high and low energy consumption locations. As a result, the amount of energy required at a high energy consumption location is shared by the two nodes instead of only one node. Hence, the lifetime of the network is significantly increased due to the mobile node rotation.

Fig.2 Mobile node rotation process

To improve the network lifetime, the sink is moved towards the last-hop relays. To moderate the relocation, a point G is defined as the equidistant position from these relay nodes in term of distancetraffic. To avoid the domination of nodes due to less traffic, the final position of the sink is fixed between its actual position and the point G. This position is determined using a dichotomy approach based on an evaluation formula. Finally, the sink is moved using a straight line movement. The repositioning of sink is accepted if the energy gain exceeds a fixed threshold. The threshold computation is performed based on the overhead generated by the sink movement. The obtained simulation results show that the sink repositioning achieves decrease in packet energy consumption, an increase in the average node lifetime, and a reduction in the transmission delay.

IV. Performance Analysis

V. Conclusion and Future Work

An Energy Aware Clustering Aggregate Node Rotation with Sink Relocation (EAC-ASR) protocol is proposed to enhance the energy efficiency of the MANET. Energy aware clustering process is done to improve the access control mechanism of the network. Aggregation of data from the nodes is performed using the data collection algorithm. Rotation of the mobile nodes and relocation of the sink are performed, to balance the energy consumption in the network, during the data transmission process. Henceforth, the energy loss occurred due to the continuous multi-hopping concept is mitigated and energy consumption due to the cluster communication is also comparatively reduced. From the simulation results, it is clearly evident that the proposed EAC-ASR protocol achieves reduction in the energy consumption and improvement in the network lifetime.

References