Efficient Cluster Based Data Collection Using Mobile Data Collector for Wireless Sensor Network

ISSN (e): 2250 – 3005 || Volume, 06 || Issue, 06|| June – 2016 ||
International Journal of Computational Engineering Research (IJCER)
www.ijceronline.com Open Access Journal Page 1
Efficient Cluster Based Data Collection Using Mobile Data
Collector for Wireless Sensor Network
Dhanya L Salian1
, Ravi B2
, Udaya Kumar K Shenoy3
1
P G Scholar, Department of CSE, NMAM Institute of Technology, Nitte, Udupi District, Karnataka, India,
2
Assistant Professor, Department of CSE, NMAM Institute of Technology, Nitte, Udupi District, Karnataka,
India
3
Professor, Department of CSE, NMAM Institute of Technology, Nitte, Udupi District, Karnataka, India
I. INTRODUCTION
Wireless Sensor Network (WSN) has been emerged as a new paradigm for information gathering in recent
years. WSN is a collection of sensor nodes which are deployed in an application area to monitor certain events
[1]. WSN plays a significant role in diverse applications such as health monitoring, emergency response,
agriculture, smart transportation, military applications etc. [2]. In spite of the application of WSNs being
diverse, most of them have a common feature i.e. collection of data at the sink. While collecting the data at the
sink, the nodes near the sink consume more energy as they receive heavy traffic. These nodes have the
responsibility of forwarding the data to the sink which results in their early depletion compared to the nodes
which are farther away from the sink. When these sensor nodes fail, the faraway sensors cannot reach the sink.
This results in disconnected network and network failure. Therefore, to collect the data efficiently without these
negative impacts is known as data gathering problem [3].
Finding an effective solution to the data gathering problem is a challenging task as it determines the network
lifetime. By establishing a well-organized data gathering scheme, the network lifetime of the WSN can be
increased. Mobility can be introduced in the WSNs to reduce the uneven energy consumption among the sensor
nodes and to solve the data gathering problem. In this paper we introduce mobility into the WSN via mobile
data collector.
In this paper, we propose a data gathering scheme which employs clustering and mobility in WSNs for data
collection. Clusters are formed and the sensor nodes send data to the cluster heads. The shortest traversal order
is calculated for the mobile data collector to visit the cluster heads in the WSN. The mobile data collector visits
the cluster heads in the calculated traversal order and collects data from the cluster heads and finally delivers the
data to the sink. The shortest traversal order saves time and energy in data gathering and hence is an effective
data gathering approach which increases the network lifetime.
The rest of the paper is designed as follows: Related work is discussed in section II. Section III discusses the
proposed scheme. Simulation results are discussed in section IV. Section V summarizes the paper.
II. RELATED WORK
Establishing energy efficient data gathering scheme has been a challenging research work and lot of researches
have been carried out to achieve this. Researches already carried out in this field are discussed in this section.
Data gathering problem is considered in the paper proposed by Ming Ma et al. [4]. Here, the mobile collector
(M-collector) visits the communication range of each sensor and collects data within a single hop
communication. The data is collected without relay. Here, the focus is on prolonging the network lifetime.
ABSTRACT
Establishing an efficient data gathering scheme in wireless sensor networks is a challenging task. Lot
of researches has been carried out to establish energy efficient data gathering scheme to avoid heavy
traffic received by the nodes near the sink. Data gathering scheme is a significant factor in
determining the network lifetime. In this paper we propose an efficient data gathering scheme by
introducing clustering and mobility into the wireless sensor network. We consider data collection in
wireless sensor networks by utilizing mobile data collector and cluster heads. Cluster heads are
chosen and clusters are formed to collect data from the sensor nodes. The proposed scheme finds the
shortest tour for the mobile data collector to collect data from the cluster heads. The shortest tour
saves time and energy in data gathering.
Keywords: Wireless Sensor Networks, Mobile Data Collector, Sink, Cluster Heads, Minimum
Spanning Tree, LEACH, Network Lifetime

Efficient Cluster Based Data Collection Using Mobile Data Collector For Wireless Sensor Network
Usage of multiple mobile data collectors and spatial division multiple access (SDMA) technique for data-
aggregation in WSNs has been introduced by Miao Zhao et al. [5]. Here, mobile collector known as SenCar is
employed in the divided non-overlapping regions of the sensing field. The data gathering problem is solved
using multiple mobile collectors and the network lifetime is increased.
Upasana Sharma et al. [3] have proposed mobile data gathering scheme which uses Mobile Data Collector
(MDC) to collect data from the cluster heads. MDC dynamically changes its gathering tour by making use of
information in a Neighbour Information Table (NIT). Cluster heads create NIT by learning about its neighbours.
MDC utilizes the information in NIT and decides its traversal tour. This paper discusses the dynamic path
reduction and data filtration in WSNs.
Controllable mobility approach is considered in the paper proposed by S. Chowdhury et al. [6]. Here, the
minimum set of sensors is chosen as Data Collection Points (DCPs). The mobile data collector visits these DCPs
in a specific sequence. Minimization of hop count and tour length of mobile data collector to conserve energy is
discussed in this paper.
Minimizing the data gathering cost has been discussed in the paper proposed by S. Guo et al. [7]. Here, the
mobile data collector collects the data at the anchor points. The mobile data collector moves in the sensing field
to collect data from the anchor points. The mobile data collector stays at the anchor point for a short duration to
gather the data forwarded by the sensor nodes. Nonuniform energy consumption among the sensor nodes is
eliminated and the data gathering latency is shortened in this approach.
III. PROPOSED SCHEME
Proposed scheme creates well organized wireless sensor network for data collection by creating clusters. Mobile
node i.e. Mobile Data Collector (MDC) is used to move and collect data from the Cluster Heads (CHs). Data is
collected from CHs at low cost via. MDC. MDC visits the CHs in the shortest path and hence it results in
efficient data collection. Fig-1 gives the basic idea of the proposed scheme. Fig-1 shows an example of shortest
tour path of MDC visiting each CHs. Sensor nodes are randomly distributed in the WSN and they result in
homogeneous network as the nodes are of similar type. It is assumed that the sensor nodes are aware of their
coordinates.
Fig 1 Proposed scheme
The goal of the proposed scheme is to increase the network lifetime and also to reduce traffic load and energy
consumption of the sensor nodes near the sink. In this scheme, initially we select the CHs among the randomly
distributed nodes using CH selection algorithm of LEACH. Next we form cluster of nodes and find the shortest
traversal path for the MDC to collect data from the CHs. The MDC then has to approach each CH in this path
and collect data from the CHs and finally transfer the data to the sink. We use a TSP solver to find the shortest
path for MDC.
1.1. LEACH Protocol
One of the clustering based hierarchical-routing protocols is Low-Energy Adaptive Clustering Hierarchy
(LEACH). Based on the threshold values the CHs are randomly selected in LEACH protocol.
1.2. CH Selection Algorithm of LEACH Protocol
Here, the nodes are randomly distributed in a network. Each node takes a decision whether to become a CH for
current round or not. Here, each node will generate a random number in the range of 0-1. If the number is less
than threshold value, then node is CH for the current round. Threshold is given by equation (1).

where Tn is the threshold value calculated for the node, Pn is the percentage of CHs in each round, R is the
current round. Pn is calculated using equation (2).
To determine the number of clusters i.e. K, heuristics used in LEACH is used. Given N number of nodes,
partition the sensor field into K uniform-sized clusters using equation (3), where K is a squared number.
Thus the CHs are selected using the CH selection algorithm of the LEACH protocol. Once the CHs are selected,
the sensor nodes form clusters by using the clustering approach which is discussed below.
1.3 Clustering approach
The clustering algorithm given below forms the cluster of nodes under the selected CHs. Each sensor nodes with
co-ordinates (x, y) calculates minimum distance using equation (4).
Sensor nodes form clusters by assigning themselves to the CHs which is at minimum distance to them. The
clusters are formed using clustering algorithm and the traffic is set between CHs and the sensor nodes. Once the
clusters are formed, the shortest traversal path for MDC is calculated.
1.4 Finding Shortest Traversal Path
The shortest traversal tour planning is required for MDC. The shortest traversal tour of MDC consists of CH
nodes to be visited. For example, let denote CH nodes selected for traversal and M be the
MDC then the traversal tour for MDC can be denoted as . The problem of finding
shortest traversal path is to determine the shortest visiting order for CH nodes.
First calculate the distance between each CH nodes and create a complete weighted graph G(Vi,Ei), where Vi is
the set of homogeneous sensor nodes selected as CHs and Ei is the set of edges between the CH nodes present in
G. Find the Minimum Spanning Tree possible for this complete graph using Prims algorithm. This MST gives a
tree with minimum weight of edges connecting any two vertices in a graph.

Applying Depth First Search to the MST we find the traversal order of CH nodes to be visited by MDC. The
MDC traverses the CHs in the calculated order and collects data from CHs and finally hands over data to the
sink.
IV. SIMULATION RESULTS
The proposed scheme is simulated using ns2. It is simulated for different topologies with 20, 40, 60, 80 and 100
nodes. The shortest traversal path is calculated in C++. Mobile node moves with a speed of 5m/s and gathers
data from the CHs. The simulation parameters used in this project are shown in Table-1.
Table 1. Simulation parameters
Parameter Values
Number of Nodes 20,40,60,80,100
Simulation area 500*500
Initial Energy of Sensor
Node
100J
Traffic Source CBR
Transmission Protocol UDP
Packet size 512 bytes
Communication range 50 meters
The simulation of proposed scheme is shown in Figure 1. The CHs selected are highlighted and even the MDC
and the sink. The network lifetime is determined for various topologies and the graphs are plotted with number
of nodes in X-axis and network lifetime along Y-axis.
Fig 2. Simulation of proposed scheme for 40 nodes
The sensor nodes are randomly distributed as shown in Fig-2.The nodes selected as cluster heads are labeled as
CH as shown in the Fig-2.
Fig 3. Number of nodes vs. network lifetime

Fig-3 shows the graph plotted for network lifetime. The threshold value is fixed and the network lifetime is
calculated for 20, 40, 60, 80 and 100 nodes. Fig-4 shows the graph plotted for the average energy consumption
of the network for 20, 40, 60, 80 and 100 nodes.
Fig 4. Number of nodes vs. average energy consumption
V. CONCLUSION
The proposed scheme achieves efficient data collection using MDC and it is more effective compared to the
static data collector. This scheme can be improvised to reduce the energy consumption between the MDC and
the sink by reducing the data transmission between MDC and the sink. Cluster based approach is used in the
proposed scheme where sensor nodes form clusters and send data to the CHs. The data aggregated at the CHs
are forwarded to the sink via MDC. The best traversal order is used by the MDC to visit the CHs to collect data.
The shortest traversal order used by the MDC saves time and energy in data collection and hence is an effective
solution for data gathering problem. The proposed scheme increases the network lifetime and reduces the load
on the sensor nodes near the sink.
REFERENCES
[1]. Mohammad A. Matin, “Wireless Sensor Networks Technology and Protocols”, InTec publications, Inc. Croatia,p. cm.ISBN 978-
953-51-0735-4,2012.
[2]. Amiya Nayak and Ivan Stojmenovic, “Algorithms and Protocols for Scalable Coordination and Data Communication”, John
Wiley & Sons, Inc. Hoboken, New Jersey, pp-153-181, 2010.
[3]. Upasana Sharma, Dr. C. Rama Krishna and Dr. T.P. Sharma, “An Efficient Mobile Data Collector Based Data Aggregation
Scheme for Wireless Sensor Networks” IEEE International Conference on Computational Intelligence & Communication
Technology, 2015.
[4]. M. Ma and Y. Yang, “Data Gathering in Wireless Sensor Networks with Mobile Collectors,” Proceedings of IEEE International
Symposium on Parallel and Distributed Processing (IPDPS), pages 1-9, 2008.
[5]. M. Zhao and Y. Yang, “Data Gathering in Wireless Sensor Networks with Multiple Mobile Collectors and SDMA Technique
Sensor Networks,” Proceedings of IEEE Wireless Communication and Networking Conference (WCNC ), pages 1-6, 2010.
[6]. S. Chowdhury and C. Giri, “Data Collection Point Based Mobile Data Gathering Scheme Relay Hop Constraint,” Proceedings of
IEEE International Conference on Advances in Computing, Communications and Informatics (ICACCI ), pages 282-287, 2013.
[7]. S. Guo and Y. Yang, “A Distributed Optimal Framework for Mobile Data Gathering with Concurrent Data Uploading in
Wireless Sensor Networks,” Proceedings of IEEE Conference on Computer Communications (INFOCOM), pages 1305-
1313, 2012.

Efficient Cluster Based Data Collection Using Mobile Data Collector for Wireless Sensor Network

Related slideshows

More Related Content

Efficient Cluster Based Data Collection Using Mobile Data Collector for Wireless Sensor Network