Maximum Lifetime Routing Strategies for Wireless Sensor Networks in Coal Mine

foamyflumpΚινητά – Ασύρματες Τεχνολογίες

21 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

59 εμφανίσεις

Maximum Lifetime Routing Strategies

for Wireless Sensor Networks in Coal Mine


Xiangping Gu
1

1
School of
Computer
Engineering

Huaiyin Institute of Technology

Huai

an
, Jiangsu Province, China

xpgu2008@gmail.com

Ying Jin
1

,
Yanjing Sun
2

2
School of Informati
on and Electrical Engineering

China University of Mining and Technology

Xuzhou, Jiangsu Province, China

Jingjie Yan
3

3
School of Information Science and Engineering

Southeast University

Nanjing,
Jiangsu Province, China


Abstract

In WSNs’ application, the

reduction of energy
consumption and the entire network’s lifetime should be
optimized and feasible communications are must be ensured.
All of the above have become to be an important factor to
evaluate the performance of routing protocols. In the paper,
w
e present a feasible routing protocol for WSNs in coal mine.
Here, we call it LEACH
-
mine. In the algorithm, the cluster
formation not only considers the geographical position factor
between the nodes but also integrates with the residual energy
of nodes to

reduce energy consumption. Moreover, in order to
ensure nodes’ end
-
to
-
end reliable transmission in coal mine,
minimum spanning tree algorithm is used to realize multi
-
hop
communications. We simulate LEACH
-
mine in NS2 and
compare it to LEACH. Simulation te
sts prove that LEACH
-
mine prolong the network lifetime.

Keywords
-
w
ireless
s
ensor
n
etworks
;

c
lustering
;

r
outing
p
rotocol
;
LEACH
;
n
etwork
l
ifetime

I.


I
NTRODUCTION


Wireless senor networks (WSNs) are composed of sensor
nodes, which measure physical parameters in
terrains of
interest, such as temperature, humidity, presence of objects
and gas concentration in coal mine etc. [1,2] and send
gathered data to a data sink where information is processed.
WSNs have been envisioned to help in numerous monitoring
applicatio
ns[3,4,5] in military,

disaster relief and civilian
fields. We can utilize it to detect, monitor and check on work
attendance; also we can use it to tracing or attempting
electric locomotive underground. As sensor nodes are left
unattended after deployment
, sensor nodes are usually
battery powered. Therefore, designing an energy efficient
routing protocol is paramount to maximum network lifetime
to meet the characteristics of coal mine.

Common wireless sensor networks consist of large
numbers of nodes, whic
h can be divided into two classes,
sink node and sensor node. In the coal mine tunnels, the
distance is remote among a great many of senor nodes and
sink node. If sink node directly in charge of these sensor
nodes, communication cost, management delay and
complexity will become the important factors affecting the
network performance[6]. One of the energy
-
efficient
techniques to extend the lifetime of a sensor network is
clustering, such as LEACH protoco
l
[7
,8
].
To extend the
network lifetime, it is often cou
pled with the data fusio
n[
9
].
Each cluster selects one node as the cluster head. The data
gathered from the sensors are forwarded to the cluster head
first, and then to the sink. The cluster heads can fuse the data
from the sensors to minimize the amount o
f the data to be
forwarded to

the sink. Clusters can be organized
hierarchically. The network coverage size and number of
nodes will vary as the exploitation of coal resources, so the
clusters method can
meet the requirement of WSNs’
scalability in coal mi
ne.

In addition
,
on the one hand,
in the
clustering routing protocol,
in order to prolong the network

s
lifetime,
we should take nodes


remnant

energy and nodes


relevant

position into consideration in the process of cluster
head election

for the complex c
oal mine
environment
; on the
other hand,
the
minimum spanning tree

algorithm
[10]
is used
to ensure nodes


reliable communication to improve the
whole network

s performance.

The rest of the paper is organized as follows. In Section 2

existing some problems
are briefly discussed as LEACH
protocol is applied to the real coal mine environment.
Section 3 describes in detail LEACH
-
mine protocol. Section
4 illustrates experimental r
esults and analyzes the novel
routing protocol performance. Finally, in section 5,
we
conclude this paper.

II.

P
ROBLEM
D
ESCRIPTIONS

A.

Cluster Head Selection

LEACH performs a periodic randomized rotation of the
cluster head to enable all the modes within the cluster to tae
on a collective responsibility in order not to drain the battery
of a si
ngle node. In the uniform string coal mine tunnels, one
clustering is adjacent to the other one. LEACH doesn’t
cons
ider

geographical position between the nodes, so the
cluster scheme is not always reasonable.
In order to reduce
the whole networks


energy c
onsumption, nodes


residual
energy have been considered in the process of cluster head
selection.
To avoid the problem, better solution should be
proposed

in LEACH
-
mine
.

B.

Inter
-
cluster Communication Method

LEACH protocol assumes that all the senor nodes can

communicate directly with the sink node. Meanwhile, both
inter
-
cluster and intra
-
cluster use single hop communication.
However, the assumptions are unreasonable in terms of
large
-
scale WSNs in the underground coal mine. Firstly, the
length of coal mine tu
nnels varies from several kilometers to
even tens of kilometers, while the range of width is from a
few meters to 10 meters, so the majority of nodes are
deployed in a parallel roadway. Sink node is left in the end
of the tunnel, so most of the cluster hea
ds are far away from
the sink node. In order to conduct long
-
distance data
transmission, it needs to follow the multi
-
path fading model,
which will lead to a substantial loss of energy and cluster
heads may soon die. However, the underground nodes
transmis
sion distance is limited. If the nodes beyond the
transmission distance, they can’t communicate with other
nodes, it may result in the loss of data or even data packets
can’t be forwarded in time. Therefore, a multi
-
hop routing
should be

used to solve the
problem.

III.

LEACH
-
MINE
P
ROTOCOL
D
ETAILS

A.

Cluster Head Selection Mechanism

Here, we call the improved protocol LEACH
-
mine.
In
the initial stage of the network, each node determines
whether it will become a cluster head or not by generated
random numbers. If the

nodes with low residual energy are
elected as cluster heads, it will bring much extra energy
consumption of cluster heads and withdraw from the
network due to forwarding the data within the cluster and
outside the clusters. It may cause the entire network

interruption, while cluster heads distribute linearly in the
underground mine tunnels. LEACH
-
mine priority considers
nodes with more residual energy as cluster head.
Specifically, we convert nodes’ residual energy into delay
criterion. Namely, if the node
’s residual energy is more than
the other, the former delay is less than the latter. In the set
-
up
phase, all the nodes need to wait for a time delay. The node
of delay priority arrival will become a cluster head in the
round, or it will become the cluste
r member node. It can
ensure the cluster heads are distributed uniformly in the
network, bring energy consumption balance and extend
overall network lifetime without performance degradation.

The cluster heads will broadcast an advertisement
containing the
TDMA time slot information. But the whole
network broadcasting is not necessary for LEACH
-
mine
protocol. In one hand, the clusters are divided according to
geographical location in the special terrain of the mine
tunnels, and different clusters are adjacen
t into a banded
linear distribution. Each cluster member node will joint the
nearest cluster, so it is not necessary to broadcast the
message in the whole network. In the other hand, the node
transmission power increases as the communication radius.
The cl
uster heads will not only consume large amounts of
energy, but will cause channels interference.

B.

Inter
-
cluster Communication Mechanism

Once the clusters are created and the TDMA schedule is
fixed, the data transmission can start. The data transmission
phas
e is

composed of two parts: inter
-
cluster and intra
-
cluster transmission. Energy consumption of the inter
-
cluster is much more than the intra
-
cluster’. Therefore, it is
obviously important how to set the reasonable inter
-
cluster
communication mechanism to
decrease cluster heads’
energy consumption. To solve the problem, the minimum
spanning tree algorithm
[11,12]
is us
ed to realize the multi
-
hop communication.

C
4
2
3
A
B

Figrue1 Cluster Head Communication Graph

Figure1 shows cluster head c
ommunication diagram. In
the diagram, A, B, C denote the cluster heads, the distances
of A and B,

A and C, B and C are 2, 4, 3 respectively. And
A and B both sent a data to C. The data fusion tree is A
-
B
-
C
in accordance with the theory of minimum spanning
tree. A
firstly sends a data to B, so its communication cost makes 2;
B will fuse received data and locally generated data, and
then sends it to C, in this case, communication cost is 3, the
total communication cost of data fusion tree is 5.

IV.

S
IMULATION EVA
LUATION

In the Linux platform, we simulate and analyze LEACH
and LEACH
-
mine in NS2. Evaluate the network
performance in terms of the network lifetime, average energy
consumption and the number of received packets of sink
node. The scene is 100m*8m, the nod
e’s initial energy is 2J,
the node’s communication range is 100m. We use the
definition of the network lifetime based on the percentage of
nodes alive in the whole network. The number of nodes is
N=70, N=100 and the proportion of the optimal cluster head
i
s p=8%.

A.

Network Lifetime

Nodes die more slowly, indicating that the algorithm can
balance the nodes’ energy consumption.


(a) Network Lifetime

(N=70)


(b)Network Lifetime

(N=100)

Figure2 Lifetime
b
etween LEACH and LEACH
-
mine

B.

Average Energy Consumption

In

terms of graph trend, figure 3 displays that LEACH
-
mine protocol performance is better than LEACH. Let’s take
N=70 for example, nodes’ average energy consumption is
0.88J and LEACH
-
mine is only 0.25J correspondingly
decreasing by 0.63J when 1000s. When 36
88s, energy in
LEACH is completely depleted. In the case, LEACH
-
mine
consumes 1.52J. By comparing with LEACH protocol, the
energy consumption declines by 24%.



(a)Average Energy Consumption (N=70)


(b) Average Energy Consumption (N=100)

Figure3 average
energy

consumption under different network sizes

C.

Number of Received Packets of Sink Node

The Figure 4 shows that the relation between the number
of received packets of Sink node and network lifetime for
LEACH and LEACH
-
mine in mine tunnel.


(a)Number of R
eceived Packets of Sink Node (N=70)


(b)Number of Received Packets of Sink Node(N=100)

Figure4 Relation between Received Packets and Lifetime

We observe that the number of received packets of sink
node in the two protocols enhances linear as the simulatio
n
time when the nodes can

t enter into the death cycle. In the
case, the extent of the enhancement in LLECH
-
mine is lower
than LEACH, which is corresponded with longer delay in
LEACH
-
mine. It is obviously that the growth rate in LEACH
is greater than LEACH
-
mine. When the nodes enter into the
death cycle, two protocols have decreased the growth rate.

V.

C
ONCLUSIONS

This paper analyzes the characteristics of the terrain in
mine tunnel and proposes the improvement strategies. As the
mine tunnel in the WSNs is a z
onal distribution and
information flow statistics are not balanced. The main
information flow transmit one
-
way leading to rod
-
bone
structure, which causes unbalanced network load. Thus, we
consider nodes’ residual energy in the cluster head election
to dec
line the network load. In addition, in the mine tunnel, it
is not necessary broadcast advertisement message for the
cluster heads, which reduces energy consumption. In the data
transmission phase, minimum spanning tree algorithm is
used to achieve long
-
dis
tance data transmission underground
mine tunnel. LEACH
-
mine has confirmed that it provides
better performance than LEACH. In future work, we will
take the into consideration multi
-
level hierarchy protocol and
also consider some factors such as the coverage
,
transmission delay to improve our algorithm.

R
EFERENCES

[1]

Kavitha.C and Dr.K.V.Viswanatha. A Hybird Reliable Routing
Technique for Wireless Sensor Network[C].IJCSNS International
Journal of Computer Science and Network Security,2009,9(3):35
-
39.

[2]

Saleh H. Al
-
Sharaeh,Ahmad A.
Multi
-
Dimensional Poisson
Distribution Heuristic for Maximum Lifetime Routing in
WSN[J].World Applied Sciences Journal.2008,5(2):119
-
131.

[3]

E.Cayirci,T.Cuplu, SENDROM :Sensor Networks for Disaster Relief
Operations Management, MedHocNet, 200
4.

[4]

T.He et al,

Energy
-
efficient Surveillance Systems Using Wireless
Sensor Networks[C].Proceeding of the ACM MobiSys, 2004.

[5]

Jiaxi You, Dirk Timmerman. Strategies for Resource Management in
WSNs[J].2009.

[6]

Mo Li, YunHao Liu.
Underground coal mine monitoring wi
th wireless
sensor networks[J].ACM Transaction on Sensor Networks.2009,5(2).

[7]

W.R.

Heinzelman, A. Chandrakasan, H. Balakrishnan, Energy
-
efficient

communication protocol for wireless micro
-
sensor
networks
[C]
,

in: Hawaii International Conference

on System Sci
ences
(HICSS), 2000
.

[8]

Siddhu Warrier.
Characterization

and Application of MANET
Routing Algorithms in WSNS[M]. Master of Science School of
Informatics University of Edinburgh,2007
.

[9]

D. Hall
.
Mathematical Techniques in Multisensor Data fusion
.

Artech
House, B
oston, MA, 1992.

[10]

Zhi Ang Eu. Routing and Relay Node Placement in WSNs Powered
by Ambient Energy Harvesting[J].2008.

[11]

G anesan D , G
ovindan R , Shenker S , et al . Highly Resilient ,
Energy

Efficient Multipath Routing in C
ommunication Review
(MC2R)
[J]

.200
2

,1 (2)
.

[12]

Deepak Ganesan.
Highly
-
resilient, energy
-
efficient multipath routing
in wireless sensor networks
[J].ACM SIGMOBILE Mobile
Computeing and Communication Review.2001,5(4):11
-
25.