The SIJ Transactions on Computer Networks & Communication Engineering (CNCE), Vol. 2, No.

1, January-February 2014
ISSN: 2321 2403 2014 | Published by The Standard International Journals (The SIJ) 5



AbstractHere, we present a new concept of M_Collectors instead of static collectors. M_Collectors provides
the mobility in the network. Main objective of mobile data-gathering scheme is to improve the scalability and
solves intrinsic problems of large-scale homogeneous networks. This new concept under the mobility provides
data gathering tour periodically from the static data sink, polls each sensor in single-hop communications, and
finally transports the data to the data sink. We mainly focus on the problem of minimizing the length of each
data-gathering tour and refer to this as the single-hop data-gathering problem. Our single-hop mobile data
gathering scheme can improve the scalability and balance the energy consumption among sensors. It can be
used in both connected and disconnected networks. Mainly focus on the problem of minimizing the length of
each data-gathering tour and refer to this as Single-Hop Data Gathering Problem (SHDGP).Our single-hop
mobile data-gathering scheme can improve the scalability and balance the energy consumption among sensors
Here we proposed a data gathering tour, by using minimum spanning tree covering algorithm to find the path
through which the M_Collector can move. Data gathering With multiple m_collector algorithm is also using
for time management and also delay can be minimized using this multiple collectors. Many limitation of static
collectors including energy consumption can be minimized by using mobility concept. Mobile collector may be
mounted upon acity buses that repeatedly follow a predefined trajectory with a periodic schedule. The use of
such existing infrastructure removes the unrealistic requirement for using dedicated mobile Controllable
platforms to carry Mobile collector. Comparison between single and multiple collectors has given in the graph
as a part of result. It gives an idea that when the number of nodes increases then the throughput of single and
multiple collectors have same increment. When the number of nodes is less then the throughput also varies. By
proposing the rendezvous node concept we can find the tour length is little greater for the single collector than
multiple collector.
KeywordsM_Collector; Static Collectors; Static Data Sink; Single-Hop; Spanning Tree.
AbbreviationsMobile Collector (M_Collector); Single Hop Data Gathering Problem (SHDGP); Wireless
Sensor Networks (WSN).

I. INTRODUCTION
UE to the recent development of technologies in
wireless communication, and in wireless sensor
networks, many new information gathering
mechanisms have developed in applications such as medical
treatment, outer-space exploration. Without using a
preconfigured infrastructure large number of sensors are
deployed in the sensing field [Bai et al., 2012]. At first these
sensor nodes must able to discover s their neighbor nodes and
form the network by themselves. Most of the energy of the
sensors is consumed on two major tasks: sensing the field and
uploading data to the data sink. Energy consumption on
sensing is stable because it does not depend on the topology,
it only depends on the sampling rate [Guo et al., 2011]. In
homogeneous network sensors are arranged in a flat
topology, sensors close to the data collector consume more
energy than sensor that are far away from the data collector
[Liang et al., 2013]. So many sensors want to relay packets
from the sensors. If any of the sensor fails then sensor cant
reach the data to the collector, then whole network will be
disconnected [Shah & S. Roy, 2003]. In those areas mobile
collector is perfectly suitable application. A mobile collector
serves as a mobile data transporter that moves through every
subnetworks. The moving path of the mobile data collector
acts as virtual links between separated sub networks.
D
*Master of Engineering, Department of Computer Science & Engineering, Kathir College of Engineering, Coimbatore, Tamilnadu, INDIA.
E-Mail: timna615{at}gmail{dot}com
**Master of Engineering, Department of Computer Science & Engineering, Kathir College of Engineering, Coimbatore, Tamilnadu, INDIA.
E-Mail: kabil444dev{at}gmail{dot}com
***Master of Engineering, Department of Computer Science & Engineering, Kathir College of Engineering, Coimbatore, Tamilnadu, INDIA.
E-Mail: chasemekavitha29{at}gmail{dot}com
Timna.P.Elizabeth*, P. Kabil Dev** & S. Kavitha***
An Efficient Data Gathering Mechanism
using M_Collectors
The SIJ Transactions on Computer Networks & Communication Engineering (CNCE), Vol. 2, No. 1, January-February 2014
ISSN: 2321 2403 2014 | Published by The Standard International Journals (The SIJ) 6
To provide a scalable data gathering scheme for large
scale sensor networks, we utilize mobile data collectors to
gather data from sensors. The mobile data collectors starts a
tour from the data sink traverses the network, collects sensing
data from near by nodes while moving, and then returns and
uploads data to the data sink [Amis et al., 2000]. The mobile
data collector moves close to sensor nodes and if the moving
path is well planned the network lifetime can be increased
[Wang et al., 2005].
II. RELATED WORKS
Here we are discussing about some related work on the data
gathering mechanism in WSNs.
In flat topology, data routing can cost energy expenditure
so introduce a hierarchy to the network. A two layered hybrid
networks are more scalable and energy efficient than
homogeneous sensor networks [Luo & Hubaux, 2005]. In
those networks sensor nodes are organized in to clusters and
form as lower layer. At higher layer, cluster heads collect
sensing data and move to the data sink. A cluster head acts as
data aggregation point for collecting sensing data from
sensors and also as a controller or scheduler to make various
routing and scheduling decisions [Somasundara et al., 2004].
In homogeneous network where all nodes have identical
capability, so some nodes are selected to serve as cluster
heads. Cluster head consume more energy than other nodes,
and it may fail faster than other nodes. So sensor nodes can
become cluster heads rotationally.
Mobility of the sensor networks has been studied
[Chessa & Santirash, 2002; Somasundara et al., 2004]. Those
studies described about radio tagged zebras and whales were
used as mobile nodes to collect sensing data in a wild
environment. These animal based nodes randomly wander in
the sensing field and exchange sensing data only when they
move close to each other. But the mobility of the random
moving animal is hard to predict and control. Thus maximum
packet delay can not be guaranteed.
A number of mobile observers called data mules pick up
data directly from the sensors when they are close in range,
buffer the data and drop off the data to the wired access
points [Shah & Roy, 2003]. The mobile observers traverse
the sensing field along straight lines and gather data from
sensors. But data mules may not be always able o move along
straight lines, for example obstacles or boundaries may block
the moving paths of data mules.
III. PRELIMINARIES
We consider the data gathering problem in which the
M_collector can visit the transmission range of every static
sensor, such that sensing data can be collected by a single hop
communication without relay.



3.1. Polling Points
While moving, an m-collector can poll near by sensors one
by one to gather data. Upon receiving a polling, message, a
sensor simply uploads the data to the M_collector directly
without relay. We define the positions where the M_collector
polls sensors as polling points [Chessa & Santirash, 2002].
When an M_collector moves to a polling point, it polls
nearby sensors within the same transmission power as
sensors, such that sensors that receive the polling message
can upload packets to the M_collector in one hop [Guo et al.,
2011]. After data gathering from sensors around the polling
point, the M_collector moves directly to next polling point in
the tour.
P={p1,p2,.pt} denotes a set of polling points
DS be the data sink.
Moving tour of the m_collector can be represented by
DS->P1->P2->.->PT->DS. So, before an M_Collector starts a
data gathering tour, it needs to determine the positions of all
polling points and which sensors it can poll at each polling
point.
IV. DATA GATHERING WITH SINGLE AND
MULTIPLE COLLECTORS
4.1. Single-Hop Data Gathering
Here, we consider the problem of finding the shortest moving
tour of an M_Collector that visits the transmission range of
each sensor. The positions of sensors are either the polling
points in the data gathering tour or within the one hop range
of polling points. In the case of travelling salesman problem
M_Collector must visit the location of every sensors one by
one to gather data. The goal of TSP is to find the minimum
distance tour that visits every points in the plane exactly
once.
4.2. Heuristic Algorithm for the Single-Hop Data
Gathering Problem
The basic idea behind our proposed Spanning Tree Covering
algorithm is to choose a subset of points from the candidate
polling point set. The algorithm tries to cover minimum
average cost at each uncovered neighbor set of sensors with
the minimum average cost at each stage.
Let Pcurr contains all polling points, L be the set of
remaining uncovered sensors at each stage of the algorithm.
Let nb(l) denote neighboring set of L. Let cost {nb(l)} be the
cost of an uncovered neighboring set.
Let $=cost{nb(l)/|nb(l)+U|}denote the average cost to
cover all uncovered sensors in nb(l).choose the nb(l) with $
minimum value.
The SIJ Transactions on Computer Networks & Communication Engineering (CNCE), Vol. 2, No. 1, January-February 2014
ISSN: 2321 2403 2014 | Published by The Standard International Journals (The SIJ) 7

4.3. Data Gathering with Multiple M_Collectors
The entire network can be divided into sub-networks. In each
sub-network, an M_collector is responsible for gathering data
from local subarea in the network [Wang et al., 2005]. From
each M_Collector the data moves to the M_collector that visit
the data sink.
Let Lmax be the upperbound of any subtour, which
guarantees the data to be collected before sensors run out of
storage [Konstantopoulos et al., 2012].
Let t(v)denote the subtree of T. Let parent(v)be the
parent of vertex v. Let weight {v}represent the sum of link
cost in the subtree t{v}.

V. MOBILE COLLECTOR MOUNTED ON
ACITY BUSES
A secret meeting-based solution and targets applications that
involve monitoring of isolated urban areas (e.g., urban parks,
building blocks, or large communal facilities) with respect to
environmental parameters, surveillance, fire detection, etc. In
such environments, Mobile collector may be mounted upon
acity buses that repeatedly follow a predefined trajectory with
a periodic schedule. The use of such existing infrastructure
removes the unrealistic requirement for using dedicated
mobile Controllable platforms to carry Mobile collector.
Mobile collector is used to collect data from groups of SNs
[Skraba et al., 2004]. During a training period, all the WSN
edge nodes located within the range of Mobile collector
routes are appointed as Polling Points and build paths
connecting them with the remainder of sensor nodes. Those
paths are used by remote nodes to forward their sensory data
to Polling Points. The proposed technique assumes full
aggregation. Apparently, this is not always possible and thus
it is rather a strong assumption. The solution presented for
fixed Mobile collector track seeks to determine a segment of
the Mobile collector track shorter than a certain bound such
that the total cost of the trees connecting source nodes with
PP is minimized.
5.1. Clustering
The large-scale deployment of WSNs and the need for data
aggregation necessitate efficient organization of the network
topology for the purpose of balancing the load and
prolonging the network lifetime [Zhao & Yang, 2012].
Clustering has proven to be an effective approach for
organizing the network in the above context [Amis et al.,
2000]. Besides achieving energy efficiency, clustering also
reduces channel contention and packet collisions, resulting in
improved network throughput under high load. Our clustering
algorithm borrows ideas from the algorithm of Chen et al., to
build a cluster structure of unequal clusters. The clustering
algorithm in constructs a multisized cluster structure, where
the size of each cluster decreases as the distance of its cluster
head from the base station increases.

Figure 1: Data forwarding paths in MobiCluster

Figure 2: Unequal Cluster Formation in MobiCluster
Data Gathering Algorithm with Multiple M_Collectors
Find the polling point set p
Find the spanning covering tree T on all polling points in
P.
For each vertex v in T, calculate the weight value
Weight(v)
WhileT#empty
Find the deepest leaf vertex u in T
Let the root of the subtree t, Root(t)=u
While Weight(parent(root(t)))<=Lmax/2
Root(t)=parent(Root(t))
Endwhile
Add all child vertices of root(t) and edges connecting
them into t and remove t from T
Update weight value of each remaining vertex in T
Endwhile
Spanning Tree Covering Algorithm
Create an empty set Pcurr
Create a set Ucurr, contain all sensors
Create a set L containing all candidate polling points
While Ucurr#empty
Find a polling point l belongs to L,
which minimizes $=cost{(nb(l))}/{nb(l)+Ucurr}
Cover sensors in nb(l)
Add the corresponding pooling points of nb(l) in to Pcurr
Remove the corresponding polling point of nb(l) from L.
Remove sensors in nb(l) from Ucurr
Endwhile
Find an approximate shortest tour on polling points in
pcurr.
The SIJ Transactions on Computer Networks & Communication Engineering (CNCE), Vol. 2, No. 1, January-February 2014
ISSN: 2321 2403 2014 | Published by The Standard International Journals (The SIJ) 8
During an initialization phase, the MC moves along its
fixed trajectory broadcasting periodically a BEACON signal
to all SNs at a fixed power level. All nodes near the MC
trajectory receive the BEACON message and thus they know
that the clusters in their region will be small sized
[Konstantopoulos et al., 2012]. Then, these nodes flood the
BEACON message to the rest of the network.

Figure 3: Comparison between Single and Multiple M_Collector




VI. CONCLUSION AND FUTURE WORK
We have proposed a mobile data gathering scheme for large
scale sensor networks. We introduced a mobile data collector
which works like a mobile station in the network. An
M_collector starts the data gathering tour periodically from
the static data sink, polls the sensors and gather the data from
sensors and finally uploads data to the data sink [Luo &
Hubaux, 2005]. Our mobile data gathering scheme improves
the scalability and solves the problems related to
homogeneous networks. By introducing M_collector, data
gathering becomes more flexible and adaptable to the
unexpected changes of the network topology. We introduced
multiple M_Collectors by allowing each of them moves
through a shorter subtour than the entire tour. Our proposed
scheme reduces the moving length, and it can prolong the
network lifetime compared with the static data collector. This
work gives a great impact in the field of data gathering in
WSN by reducing the energy consumption of each sensors
and there by increase the network lifetime.


Algorithm: Data_ Distribution
D: amount of data available at u for distribution among
the RNs attached to u
sort the RNs in R
u
in Comp
val
decreasing order into a
sorted list v1, v2,.,v
|R
v
|
(V
j
.Comp
val
<v
i.
Comp
val,
there exist, j1<i<j<|Ri
v
|)
i=0
while(D>0 and |R
v
| > i+1)
calculate:
l
i =
v
i,
T
last
- v
i.
T
first
; D
I
= r
I
. l
i

D = D D
I=i+1;
End while
Algorithm: Polling Point Node Selection
Initialize n
b
= 0, v. T first = 0, v. T
last
=0
Wait until a BEACON is received
Record BEACON receipt time t1 and single strength s1
v. T.
first
= t1, v.T
last
=t1, n
b
= 1, n
b, r
= 1
start Conection Dropped Timer
while Coneection Dropped Timer has not expired
wait until next BEACON is received or Connection
Dropped Timer is expired
If a BEACON I is received
Record BEACON receipt time t
i
and signal
Strength s
i

n
b
= n
b
+ |t
i
-t
i
-1| / |T
beacon
|
n
b,r
= n
b,r
+ 1
v. T
last
= t
i

reset Connection Dropped Timer
end if
end while
v.Comp
val
= a1 . ((E
residual
) / (E
max
) ) + a2.nb + a3 .((sum
(i=1 to nb)) / (nb) s
i

Send
RN_Cand_Msg(v.Node_I D, v.Comp
val
, v.T
first
, v.T
last
)
The SIJ Transactions on Computer Networks & Communication Engineering (CNCE), Vol. 2, No. 1, January-February 2014
ISSN: 2321 2403 2014 | Published by The Standard International Journals (The SIJ) 9
REFERENCES
[1] A.D. Amis, R. Prakash, T.H.P. Vuong & D.T. Huynh (2000),
Max-Min D-Cluster Formation in Wireless Adhoc Network,
Proceedings of IEEE Nineteenth Annual Joint Conference of
the IEEE Computer and Communications Societies, Pp. 3241.
[2] W.R. Heinzelman, A. Chandrakasan & H. Balakrishnan (2000),
Energy Efficient Communication Protocol for wireless
Microsensor Networks, Proceedings of the 33rd Annual
Hawaii International Conference on System Sciences.
[3] A. Mainwaring, J. Polastre, R. Szewezyk, D. Culler & J.
Anderson (2002), Wireless Sensor Networks for Habitat
Monitoring, ACM International Workshop on Wireless
Network Applications.
[4] S. Chessa & P. Santirash (2002), Crash Fault Identification in
Wireless Sensor Networks, Computer Communications, Vol.
25, No. 14, Pp. 12731282.
[5] R.C. Shah & S. Roy (2003), Data Mules: Modeling a Three-
Tier Architecture for Sparse Sensor Networks, Proceedings of
the First IEEE International Workshop on Sensor Network
Protocols and Applications, Pp. 3041.
[6] P. Skraba, H. Aghajan & A. Bahai (2004), RFID Wakeup in
Event Driven Sensor Networks, SigComm Poster Session,
Portland, OR, Aug. 2004. [Online]. Available:
http://conferences.sigcomm.org/sigcomm/2004/posters.
[7] A.A. Somasundara, A. Ramamoorthy & M.B. Srivastava
(2004), Mobile Element Scheduling for Efficient Data
Collection in Wireless Sensor Networks with Dynamic
Deadlines, Proceedings of 25th IEEE International Real-Time
Systems Symposium, Pp. 296305.
[8] W. Wang, V. Srinivasan & K.C. Chua (2005), Using Mobile
Relays to Prolong the Lifetime of Wireless Sensor Networks,
Proceedings of the 11th Annual International Conference on
Mobile Computing and Networking, Pp. 270283.
[9] J. Luo & Hubaux (2005), Joint Mobility and Routing for
Lifetime Elongation in Wireless Sensor Networks,
Proceedings of 24th Annual Joint Conference of the IEEE
Computer and Communications Societies, Vol. 3, Pp. 1735
1746.
[10] L. Guo, R. Beyah & Y. Li (2011), SMITE: Astochastic
Compressive Data Collection Protocol for Mobile Wireless
Sensor Networks, Proceedings IEEE INFOCOM, Pp.1611
1619.
[11] C. Konstantopoulos, G. Pantziou, D. Gavalas, A.
Mpitziopoulos & B. Mamalis (2012), A Rendezvous-based
Approach Enabling Energy-Efficient Sensory Data Collection
with Mobile Sinks, IEEE Transactions on Parallel and
Distributed Systems, Vol. 23, No. 5, Pp. 809817.
[12] F. Bai, K.S. Munasinghe & A. Jamalipour (2012), A Novel
Information Acquisition Technique for Mobile-Assisted
Wireless Sensor Networks, IEEE Transactions on Vehicular
Technology, Vol. 61, No. 4, Pp. 17521761.
[13] M. Zhao & Y. Yang (2012), Bounded Relay Hop Mobile Data
Gathering in Wireless Sensor Networks, IEEE Transactions
on Computers, Vol. 61, No. 2, Pp. 265277.
[14] W. Liang, P. Schweitzer & Z. Xu (2013), Approximation
Algorithms for Capacitated Minimum Forest Problems in
Wireless Sensor Networks with a Mobile Sink, IEEE
Transactions on Computers, Vol. 62, No, 10, Pp. 19321944.
Timna.P.Elizabeth: I am doing M.E., in
department of Computer Science & Engineering
at Kathir College of Engineering, Coimbatore. I
graduated in B.Tech., in Computer Science &
Engineering from Sree Narayana Gurukulam
College of Engineering during the year 2012. I
have presented 4 papers in international
conferences and three national level
conferences conducted at different colleges.