Adv. Sci. Lett. RESEARCH ARTICLE [631252]

Adv. Sci. Lett. RESEARCH ARTICLE

1

Routing approach in Wireless Sensor
Network
Abdelmalek BOUDRIES 1,* , Mourad AMAD 2, Rabah KASSA 1
1 Laboratory LMA, University of Bejaia, Algeria
2 Laboratory LAMOS, University of Bouira, Algeria

In this paper, we propose a routing approach in wir eless sensor network by taking into account the con nectivity maintenance
between sensor nodes. Each node contains a weight c alculated according to its remaining energy rate. D uring the routing
process, if the transmitting node receives an updat e package of a node participated in the routing, th en it re-examines its
routing way choice, by comparing the weight of the updated routing way with the weight of the routing ways received at the
time of its search of a way for the routing in its routing table. An example analysis and simulation s how that the proposed
approach is effective.
Keywords: Connectivity, Sensors networks, Routing protocols, Fa ilure node, Energy constraint

1. INTRODUCTION
Wireless sensor network is a set of devices that ca n
monitor and collect data in a given environment, pr ocess
and transmit them to a base station where the appli cation
is located. Ensure direct communication between the
sensor nodes and the base station forces them to ma ke
their messages with high power [1] and thus cause the
depletion of their essential resource which is the battery.
Therefore, the cooperation of nodes to ensure that the
remote nodes communicate with the base station is
required. Thus, from a node to another, messages ar e
propagated forming a multi-hop route from source no de
to the base station. Routing is the process determi ning the
path between the source and destination for data
transmission [2] .
The lifetime in a wireless network, especially a
network of wireless sensors depends heavily on the
connectivity factor between its nodes. A number of factors
can be at the cause of a break in connectivity such as lack
of energy at an important node. The failure of a se nsor
node leaves some or all of its area without coverag e, and
can result the isolation of one or more sets of nod es of the
entire network, if it is a relay node. In other cas es may
cause a failure of the mission assigned to the netw ork.
Being given wireless sensors network deployed in a hard
area difficult to access. Minimize the energy consu mption
of sensor nodes due to the delivery of data towards the
base station, and the sharing of energy needed to
accomplish this task, is a good way to delay the no des
failure due to lack of energy of their battery, and prolong
the lifetime of the entire network. Our goal is to share the total energy consumed, to route data towards the ba se
station, from multiple nodes to minimize the indivi dual
energy consumption.
In this paper, we propose a novel approach for
prolonging the lifetime of the wireless sensor netw ork by
minimizing the consumed nodes energy due to the dat a
routing, in order to continue to provide measuremen t
functions or data routing. The rest of this paper i s
structured as follows: in Section II, we present re lated
works. Section III devotes our proposed Max Weight
Protocol, followed by Section IV which presents our
approach evaluation and shown its efficiency. Final ly, in
Section V, we conclude and we state prospects.

*Email Address: abdelmalekboudries@gmail.com

2. RELATED WORK
Several works are assumed to maintain connectivity in
wireless sensor networks. We can classify the appro aches
listed in the literature in two classes: curative
maintenance and preventive maintenance. The authors of
the first class work on the connectivity maintenanc e in
sensor node failure case. In [3, 4, 5, 6] , authors try to
solve the problem of connectivity, in the case of t he
network partitioning, and/or the covering of the
monitoring zone problem. The authors of the second class
work to keep the sensors network connected longest time
possible; they are looking for solutions to extend the
lifetime of the network in question. As an example, works

Adv. Sci. Lett. RESEARCH ARTICLE

2 [7, 8, 9, 10] tries to use, at a given time, a minimum of
sensors ensuring the connectivity and/or the networ k
coverage, the other sensors are put in sleeping mod e. The
solution in this second class can also be conceived at the
network deployment, to see the work of N. Ait saadi and
all [11], the goal in this solution is to generate the best
network topology while minimizing the cost of
deployment, to guarantee the maintenance of connect ivity
and the monitoring value of the coverage zone, but also to
optimize the lifetime of the network. Always in the
second class, Liao and Wang [12] propose an
asynchronous MAC protocol (AMAC), they expect to
improve the problem of energy dissipation and time
management due to sleeping program exchange under t he
PMAC basic protocol. Routing protocols also play a very
significant role to prolong the network lifetime; t he paths
for data transfer are selected in such a way that t he total
energy consumed along the path is minimized [13] .
Effective routing in a sensor network requires need to
minimize the energy dissipation. To extend the net work
lifetime, R.V. Biradar and all in [14] have realized several
multihop flat based routing protocols; and in [15] Liu and
Wang develop a maximizing energy utilization routin g
protocol. To minimize energy dissipation and maximi ze
network lifetime, Chamam and Pierre propose, in [16] , a
new distributed clustering algorithm. In their solu tion,
cluster heads are elected following three way messa ge
switch between each sensor and its neighbors. They show
its superiority (in terms of network lifetime and r atio of
elected cluster heads) in opposition to EESH, one o f the
most recent cluster algorithms [17] .
Table I. Routing protocols for WSNs

Routing in WSNs differs from traditional routing in fixed
networks in various ways. There is no infrastructur e, the
wireless links are not very reliable because sensor nodes can
fail, and routing protocols must meet strict energy -saving requirements [18] . Various algorithms of routing were
developed for wireless networks in general. All the main
routing protocols proposed for WSNs can be divided into
several categories . Akkaya and Younis in [19] present a
classification for the various approaches, they div ided
them in three main categories which are: data-centr ic,
hierarchical and location-based. Singh and all [20]
propose a classification based on seven categories as
shown in Table I.
3 MAX WEIGHT PROTOCOL
Max Weight is a routing protocol that takes into
account the energy constraint and network connectiv ity
by sharing the consumed energy for routing data rep orted
from a given sensor node to a base station, by seve ral
nodes. The idea of sharing the consumed energy by
multiple nodes involves minimizing the consumed ene rgy
individually, i.e. extend the lifetime of each node . This
leads to maintain network connectivity and extends the
lifetime of the entire network.
Our routing protocol called MWP (Max Weight
Protocol) is described in the following steps:
1. Transmitting data to the sink is achieved based on a
broadcast of a hello_sink message;
2. At the reception of the hello_sink message by node n, ,
in its turn, it broadcasts it (sent to all its neig hbors
except the one who sent it in order to eliminate cy cle);
3. When the sink receives the message, it responds by
broadcasting the hello_response message containing
the weight field = 0;
4. If a node n receives a hello_response message, it saves
the some information (source, destination and weigh t)
for later use. Adds its identifier and adds its wei ght to
the weight field, then it diffuses the message;
5. When the transmitter node of hello_sink message
receives hello_response messages from the sink, it
compares weight fields messages and selects the
largest path according to : 1 – r hops_numbe weight ;
6. If the node energy decreases, so that its weight
changes then it recalculates and adds it to the sav ed
weight (see step 4). Then it sends it in an update
message, using the routing path, to the transmitter
node which will decide to change or not the routing
path in real time;
7. If a node receives an update message then it calcul ates
and adds its weight to the update message weight an d
sends it using the routing path, to the routing sou rce
node;
8. If the transmitter node receives the update message
from a node n when routing then he sees his choice of
routing path.
After choosing the routing path, the source node st arts
sending data packet by packet; when the weight of a given
node, of routing path, decreases, this latter sends an update
message by the reverse path routing, this update me ssage
will be routed (see step 7) until arrive at the sou rce which Category Representative Protocols
Location-based
Protocols MECN, SMECN, GAF, GEAR, Span, TBF,
BVGF, GeRaF
Data-centric
Protocols SPIN, Directed Diffusion, Rumor Routing,
COUGAR, ACQUIRE, EAD, Information-
Directed Routing, Gradient-Based Routing,
Energy-aware Routing, Information-Directed
Routing, Quorum-Based Information
Dissemination, Home Agent Based Information
Dissemination
Hierarchical
Protocols LEACH, PEGASIS, HEED, TEEN, APTEEN
Mobility-based
Protocols SEAD, TTDD, Joint Mobility and Routing,
Data MULES,
Dynamic Proxy Tree-Base Data Dissemination
Multipath-based
Protocols Sensor-Disjoint Multipath, Braided Multipath,
N-to-1 Multipath Discovery
Heterogeneity-
based Protocols IDSQ, CADR, CHR
QoS-based
protocols SAR, SPEED, Energy-aware routing

Adv. Sci. Lett. RESEARCH ARTICLE

3 will decide to change or not the routing path compa ring
this new path to the paths stored in its routing ta ble.
A. Weight calculation
We define the weight calculation of a node n based on
the percentage of its remaining energy as in formul a 1:
Weight = µ x p (1)
Such as:
P: is the percentage of the energy of n; and

<<≥≥
=
30 %50 30 %50 %
energy its of the if and energy its of the if energy its of the if
u
γβα

α, β and γ are experimental variables.
We give: α = 1, β = 3/4, γ = ½
The functional principle is that if the rate of the
remaining energy decreases then both the weight of the
node and the weight of the path composing this node
decreases. The idea is to maintain as long as possi ble, the
network connected by extending the lifetime of sens or
nodes that have consumed a lot of energy (more than half
of the energy of their battery).
B. Scenario example
In the example illustrated in Figure 1, to go from node f
to node a and not to deplete the sensor node c, the sensor
node f will choose the path 1 ( f-d-b-a) that the path 2 ( f-e-
c-a) so that:

Figure 1. Example of the weight calculation of
sensor nodes
Each node calculates its weight, we have:
• weight (b) = 1 x 60 = 60
• weight (c) = 1/2 x 20 = 10
• weight (d) = 1 x 55 = 55
• weight (e) = 1 x 100 = 100
The weight of the path 1 = 55 + 60 = 115
The weight of the path 2 = 100 + 10 = 110
The sensor node f sends its data via the path 1 because
the corresponding weight is greater than the weight of the
path 2.
The choice falls on the path 1, although in the pat h 2,
there is the node e which possess the totality of its energy,
but the node c has only 20% of its energy, so extending
the lifetime of this node returns to extend the lif etime of
the path 2 and maintain the link redundant between the
node f and the node a which is very important in terms of connectivity in order to extend the global network
lifetime.
4 PERFORMANCES EVALUATION
In order to evaluate our proposed model, we take th e
network illustrated in figure 2, and we assume the
following:
• The nodes energy percentage in Figure 2 is
described in Table II;
• Each sending and receiving of 100 packets
consumes one unit of energy during one unit of
time (second).
• S is a source node and D is the destination node

Figure 2. Network example
Table II. Nodes energy percentage located between
the nodes S and D.

The process is stopped when one node consumes all i ts
energy.
A. Using our proposed protocol called Max Weight
Protocol (Table III):
We define the denotation bellow:
• W: Weight
• HN: Hops number
• SP: Sent packets
• CER: Consumed energy rate
• RER: Remained energy rate
• NW: New weight
• T: Time
Table III. Data transfer details between the nodes S
and D.

The total number of data packets sent from the sour ce
node S to destination node D is equal to 28300 pack ets
before the failure of the first node.
The required time to send these packets is 354 sec. Path 1 Path 2
f b energy =60% c energy =20% a
d energy =55% e energy =100% D
n2
S n3 n1 n5
n4
Node %_energy
n1
n2
n3
n4
n5 80
100
90
100
100

Path W HN SP CER RER NW T
S_ n2_D
S_n4_n5_D
S_n3_D
S_n1_D
S_n1_D
S_n2_D
S_n3_D
S_n4_n5_D
S_n1_D 100
100
90
80
36,75
36,75
36,75
36,75
14,5 2
3
2
2
2
2
2
3
2 5100
5100
4100
3100
2000
2000
2000
2000
2900 51
102
41
31
20
20
20
40
29 49
98
49
49
29
29
29
58
0 36,75
36,75
36,75
36,75
14,5
14,5
14,5
14,5
0 51
102
41
31
20
20
20
40
29

Adv. Sci. Lett. RESEARCH ARTICLE

4 B. T he results for each path taken without the upd ate
(Table IV):
Table IV. Data transfer details between the nodes S
and D for each path.
Path W HN SP CER RER NW T
S_ n 2_D 100 2 10000 100 0 0 100
S_n 4_n 5_D 100 3 10000 200 0 0 200
S_ n 3_D 90 2 9000 90 0 0 90
S_ n 1_D 80 2 8000 80 0 0 80
The maximum number of data packets sent from the
source node S to destination node D without updatin g the
routing path during the sending is equal to 10000 p ackets
before the failure of the first node.
Table V shows a summary of the obtained results in
Tables III and IV:
Table V. Summary of the obtained results
Path Sent_packets_nbr Time
Max Weight Protocol 28300 354
Unique path during the data transfer
S_n 2_D 10000 100
S_n 4_n 5_D 10000 200
S_n 3_D 9000 90
S_n 1_D 8000 80
We note that the report between the number of packe ts
sent and the elapsed time before the failure of the first
node is better with Max Weight Protocol than with a ny
other possible paths, this reinforces our solution in terms
of connectivity and the lifetime of the entire netw ork.
Considering the following assumptions:
– Number of nodes = 100
– Nodes power = 100 units
– Energy of trans/recep of 1 data packet = 1 mu
– During the simulation, the source and destination
are already the same. The process stops as soon as
the first node consumes all of its energy.
The results of the simulation are shown in figures 3, 4, 5
and 6.
Figure 3 illustrates a comparison of the Number of
packets sent according to the consumed energy rate for the
proposed and the basic approach (without updating).
050000 100000 150000 200000 250000 300000
50,001 70,001 100
Consumed energy rate (%) Sent packets MWP
Without updating

Figure 3. Number of packets sent according to the
consumed energy rate
We note that in Figure 3, the number of packets sen t
using updating route during routing (our Max Weight
Protocol) is much larger than without updating what ever
the route used for routing and this for the same pe rcentage
of consumed energy. The difference of packets sent
between these two approaches is illustrated in Figu re 4. 050000 100000 150000
50,001 70,001 100
Consumed energy rate (%) Difference of sent packets

Figure 4. Difference of sent packets before the
failure of the first node between updating and
without updating route protocol according to the
consumed energy rate
Figure 5 illustrates a comparison Time elapsed befo re
the failure of the first node according to the cons umed
energy rate for both proposed approach and basin on e
(without updating)
0500 1000 1500 2000 2500 3000
50,001 70,001 100
Consumed energy rate (%) Time (sec) MWP
Without updating

Figure 5. Time elapsed before the failure of the first
node according to the consumed energy rate
Figure 5 shows that the time elapsed before the fai lure
of the first node using updating route during the r outing
(our Max Weight Protocol) exceeds the time elapsed
before the failure of the first node whatever the r oute used
for routing without updating, for the same percenta ge of
consumed energy. The time difference before the fai lure
of the first node between the two approaches is ill ustrated
in Figure 6.
0,00 500,00 1000,00 1500,00
50,001 70,001 100
Consumed energy rate (%) Time difference (sec)

Figure 6. Time difference before the failure of the first
node between updating and without updating route
protocol according to the consumed energy rate
5 CONCLUSION
Maintaining a connected inter-node topology is very
crucial in mobile sensor networks applications. A f ailure
of a node can cause the network to partition and th us
disrupts the application assigned to a network. In other
words, some nodes can be disconnected from the glob al
network. Consequently, resulting parts will be
disconnected. The wireless sensors networks are gen erally
deployed in hard and difficult access environments where

Adv. Sci. Lett. RESEARCH ARTICLE

5 the breakdowns or failures of sensors nodes are pos sible.
These nodes failures can harm the connectivity of t he
entire network. Ensure direct communication between the
sensor nodes and the base station forces them to ma ke
their messages with high power and thus cause the
depletion of their essential resource which is the battery.
Therefore, the cooperation of nodes to ensure that the
remote nodes communicate with the base station is
required. Thus, messages are propagated from a node to
another, forming a multi-hop route from source node to
the base station.
In this paper, we propose a routing approach taking
into account the connectivity maintenance in wirele ss
sensor networks. Each node contains a weight calcul ated
according to its remaining energy rate. During the routing
process, if the transmitting node receives an updat e
message, of a node participated in the routing, the n it re-
examines its routing way choice by comparing both
weight of the updated routing way and the weight of the
routing ways, received at the time of its search of a way
for the routing in its routing table. Performance e valuation
shows that the results as globally satisfactory. In our
proposed routing approach, the sensors nodes consum e
less energy. The optimized values of α, β and γ in formula
1 is an important problem that should be discussed in
future work. In addition as a perspective, we will integrate
and compare our solution with several protocols to better
evaluate the performance of this solution.
6 REFERENCE
[1] L. J. G. Villalba, A. L. S. Orozco, A. T. Cabr era and
C. J. B. Abbas; Routing Protocols in Wireless Sensor
Networks ; Sensors 2009, 9, 8399-8421; doi:10.3390/
s91108399; ISSN 1424-8220.
[2] P. Kumar, M.P.Singh and U.S.Triar; A Review of
Routing Protocols in Wireless Sensor Network ;
International Journal of Engineering Research &
Technology, ISSN: 2278-0181, Vol. 1 Issue 4, 2012.
[3] A. A. Abbasi, M. Younis, and K. Akkaya; Movement-
Assisted Connectivity Restoration in Wireless Senso r
and Actor Networks ; In IEEE Transactions on parallel
and distributed systems, VOL. 20, No. 9, September
2009, pp. 1366-1379.
[4] N. Tamboli, M. Younis; Coverage-aware connectivity
restoration in mobile sensor networks ; Journal of
Network and Computer Applications, Vol. 33, Issue
4, 2010, pp. 363-374.
[5] A. Zamanifar, O. Kashefi and M. Sharifi; AOM : An
efficient approach to restore actor-actor connectiv ity
in wireless sensor and actor networks ; In
International Journal of Computer Networks &
Communications, Vol.1, No.1, 2009.
[6] A.M. Boudries, M. Aliouat, P. Siarry; Detection and
replacement of a failing node in the wireless senso rs
networks ; In Computer and Electrical Engineering
Journal, Vol. 40, Issue 2, pp. 421-432, 2014. [7] A. Makhoul, R. Saadi, C. Pham; Surveillance vidéo
sur réseaux de capteurs sans-fils: ordonnancement
adaptatif avec prise en compte de la criticité ; 14e
CFIP, Strasbourg, France, October 2009.
[8] D. Tian and N. D. Georganas; A coverage-preserving
node scheduling scheme for large wireless sensor
networks ; In WSNA’02: Proceedings of the 1 st ACM
international workshop on Wireless sensor networks
and applications. NY, USA: ACM, 2002, pp. 32-41.
[9] F Ye, G Zhong, S Lu and L Zhang; Peas: A robust
energy conserving protocol for long-lived sensor
networks ; In The 23nd International Conference on
Distributed Computing Systems (ICDCS), 2003.
[10] H. Zhang and J. C. Hou; Maintaining sensing
coverage and connectivity in large sensor networks ;
Ad Hoc and Sensor Wireless Networks journal
(AHSWN), pp. 89-123, 2005.
[11] Nadjib Aitsaadi, Nadjib Achir, Khaled Bousset ta,
Guy Pujolle; Artificial potential field approach in
WSN deployment : Cost, QoM, connectivity, and
lifetime constraints ; In journal of Computer
Networks, Vol.55, No.1, pp. 84–105, 2011.
[12] Wen-Hwa Liao, Hsiao-Hsien Wang; An
asynchronous MAC protocol for wireless sensor
networks ; Journal of Network and Computer
Applications 31 (2008), pp. 807–820
[13] V. Kumar, S. Jain, and S. Tiwari, Energy Efficient
Clustering Algorithms in Wireless Sensor Networks:
A Survey , International Journal of Computer Science
Issues, Vol. 8, Issue 5, No 2, 2011; pp. 259-268.
[14] R. V. Biradar, S. R. Sawant, R. R. Mudholkar, and
V.C. Patil, Multihop Routing In Self-Organizing
Wireless Sensor Networks , IJCSI International
Journal of Computer Science Issues, Vol. 8, Issue 1 ,
2011; pp. 155-164.
[15] Y. Liu, Z. Wang, Maximizing energy utilization
routing scheme in wireless sensor networks based on
minimum hops algorithm , Computers and Electrical
Engineering 38 (2012) 703–721.
[16] A. Chamam, S. Pierre, A distributed energy-efficient
clustering protocol for wireless sensor networks ,
Computers and Electrical Engineering 36 (2010) 303–
312 .
[17] I. Woungang, Wireless ad hoc, sensor and mesh
networks , Computers and Electrical Engineering 36
(2010) 259–260.
[18] S. Misra et al. (eds.); Guide to Wireless Sensor
Networks ; Computer Communications and Networks,
DOI: 10.1007/978-1-84882-218-4 4, Springer-Verlag
London Limited 2009.
[19] K. Akkaya, M. Younis; A survey on routing
protocols for wireless sensor networks ; Ad Hoc
Networks 3 (2005) 325–349.
[20] S. K. Singh, M. P. Singh, and D. K. Singh; Routing
Protocols in Wireless Sensor Networks – A Survey ;
International Journal of Computer Science &
Engineering Survey (IJCSES) Vol.1, No.2, 2010.