Energy efficient routing algorithm in wireless sensor networks

Computer Engineering and Intelligent Systems www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.4, No.5, 2013
60
Energy Efficient Routing Algorithm in Wireless Sensor Networks
Maryam El azhari1*
Ahmed Toumanari2*
Rachid Latif3
ESSI, National School of applied Sciences, Ibnou zohr university, Agadir Morocco
* E-mail of the corresponding author: maryam.ensa@gmail.com
Abstract
This Wireless sensor network (WSN) is widely considered as one of the most important technologies for
the twenty-first century, it provides the availability of small and low-cost sensor nodes with the ability of
sensing different types of physical and environmental conditions, data processing, and wireless
communication. Sensor nodes have a limited transmission range, and their processing and storage
capabilities as well as their energy resources are also limited. Thus, optimized routing algorithms for
wireless sensor networks should be utilized in order to maintain the routes in the network and to ensure
reliable multi-hop communication under these conditions.
Keywords: Wireless Sensor Networks, Energy Efficiency, Routing Protocols, Solar Sensors, Mobile Agents
1. Introduction
In the last years, wireless sensor networks (WSNs) have gained increasing attention from both research
community and actual users. Since sensor nodes are battery-powered devices, the main problematic to resolve is
to minimize the energy consumption of nodes so that the network lifetime can be extended to reasonable periods
of time (Brussel et al.1998). Several algorithms were proposed to meet strict energy saving requirements which
may be divided into: Location; Data-centric; Mobility; Multipath; Heterogeneity; Qos; and other specific
criterions (Chiara Buratti et al.2009). However, those protocols based on the client-server concept don’t actually
consider the fact that a sensor node has to afford various capabilities for multiple applications. That’s why
researches introduced the mobile agent (MA) approach. The Mobile Agent saves a great amount of energy but
still presents much more latency compared with the client –server approach. Looking for alternative energy
source seems efficient for prolonging the network lifetime moreover, solar energy has become more attractive
recently because of its environmental benefits and because the efficiency of photovoltaic cells has increased
significantly in the past few years, that’s why, in this work, we tried to find a trade-off balance between the
energy consumption and task latency using both solar powered nodes and mobile agents and we presented
afterwards the SMA algorithm (solar aware routing with the mobile agent concept ) which is a new routing
algorithm for energy efficiency based on solar powered sensors as alternative energy source. The remainder of
this paper is organized as follows: in Section 2 and 3 we highlighted the advantages of using the mobile agent
and solar power nodes. In Section 4 we present the proposed algorithm for energy efficiency in WSNs and finish
up with perspectives of our work.
2. Mobile Agent
The application-specific nature of tasking a wireless sensor network (WSN) requires that sensor nodes have
various capabilities for multiple applications. It would be impractical to store in the local memory of embedded
sensors the entire program needed to run every possible application, due to the tight memory constraints. A
mobile agent (MA) is a special kind of software that migrates among network nodes to carry out task(s)
autonomously and intelligently in response to changing conditions in the network environment, in order to
achieve the objectives of the agent dispatcher (Min Chen et al.2006). There are seven good reasons for using
Mobile agents: Secure Brokering distributed information retrieval, parallel processing information dissemination,
monitoring and notification, workflow applications and groupware, telecommunication networks services (Corke
et al.2010). The use of MAs to dynamically deploy new applications in WSNs has been proven to be an effective
method to address this challenge. Recently there has been a growing interest on the design, development, and

Vol.4, No.5, 2013
61
deployment of MA systems for high-level inference and surveillance in WSNs. The agent design in WSNs is
decomposed into four components, i.e., architecture, itinerary planning, middleware system design and agent
cooperation. Among the four components, itinerary planning determines the order of source nodes to be visited
during agent migration, which has a significant impact on energy performance of the MA system. It has been
shown that finding an optimal itinerary is NP-hard. Therefore, heuristic algorithms are generally used to compute
competitive itineraries with a sub-optimal performance.
3. Using Solar Powered Nodes for Power Efficiency
Solar energy has become more attractive recently because of its environmental benefits and because the
efficiency of photovoltaic cells has increased significantly in the past few years. Solar energy is derived from
Nature’s greatest renewable resource and it is non-toxic in nature (Xiaoping Fan et al.2007). Solar cells can be
utilized to power the sensor as well as to charge the batteries for WSNs, where their design goal is to provide
autonomy lifetime (for wireless sensor nodes), making WSNs more valuable in terms of versatility and longevity.
Solar cells are well-developed technology compared to other energy harvesting techniques (Jean Bulking et
al.2008), they are fairly compact, capable of efficiently generating multiple potentials, and compatible with other
processors. Two generations of solar cells are distingued:
3.1 Belk solar cells
Figure 1 shows the Bulk solar cells, they are better suited to high light conditions and the outdoors spectrum,
with efficiency ~ 15% - 20%; the poly-crystalline Si is cheaper, but less efficient than single-crystalline silicon.
3.2 Thin-film solar cell
Figure 2 shows thin-film solar cells. The spectral response is more closely matches that of artificial indoor light;
amorphous Si presents conversion efficiencies are ~5-10%; CdTe cells with good PV potential that give rise to
environmental concern
Figure 1. Bulk solar cells structure Figure 2. Thin-film solar cells component

Vol.4, No.5, 2013
62
Freie Universität Berlin laboratory team has highlighted the pros of solar powered nodes by using sensor board
hardware (Wei cai et al.2011.), it consists of a Texas Instruments MSP430 controller as core and a set of
associated sensor hardware. These sensors are:
• A light sensor for the detection of visible light
• A passive infrared sensor for detection of movement
• A temperature sensor
• A gravitation sensor for the detection of movement of the sensor board
• A microphone for determination of the ambient noise level
Figure 3. Sensor board with attached solar cells
4. The Proposed Algorithm
In this section, we present a routing algorithm for energy efficiency in WSNs christened SMA algorithm which
consists of two phases: set-up phase and the steady phase. Our algorithm assumes that the sink node knows the
geographic information of all source nodes and that the primary itineraries planning are executed at the sink
Node with unlimited energy resources.
4.1 The Set-Up Phase
The main purpose of the set-up phase is to find the nearest auxiliary sink (Solar Node) to sensor nodes, which
consist of calculating the matrix of minimal distances A:
A = [di,j]
di,j = min(dist(Sensor Node, SinkNodek) )
Where:
j ϵ {1.....n}; k,i ϵ {1….m}; m=∑Sink Nodes ; (1)
n= max (Size(Ai))
After calculating the minimal distance matrix, the sink node will then solve the shortest path of each entry of the
matrix from the sink nodes to the source nodes using Dijkstra’s algorithm (indicate the source of Dijkstra
algorithm) and then send the corresponding paths to the auxiliary nodes as show in Figure 4.

Vol.4, No.5, 2013
63
Figure 4. Setup phase of SMA algorithm Figure 5. Steady phase of SMA algorithm
4.2 The Steady State
After receiving the short paths, each solar node (i.e sink node) will generate a mobile agent that will
circulate the itinerary e.g. the minimum path in order to gather data and send it back to the sink node as
shown in Figure 5.The solar nodes will then communicate the gathered data directly to the sink node,
therefore a great amount of energy is saved, the end-to-end delay is reduced, and EPD converges
considerably to smaller values.
4.3 SMA Execution Steps
SMA algorithm starts with the set-up phase ad mentioned in Figure 6, followed by the steady phase. If the
location of a sensor is changed or its battery is running out of energy, the setup phase is taking place again in
order to recalculate the new paths otherwise the same old paths are used.
Figure 6. SMA algorithm execution steps
4.4 Simulation Steps
We implemented the basic of SMA algorithm in NetTopo environment and we compared it with both

Vol.4, No.5, 2013
64
CSWSN and MAWSN presented in (Min Chen et al.2006) based on the following metrics: energy
consumption, average end-to-end delay and cost of communication (energy *delay) .We determined
afterwards the main criteria under which the SMA is more efficient than CSWSN and MAWSN. A set of
parameters was considered to highlight the tendency of the metrics such as: target source number (nsource),
reduction ratio(r), size of sensed data at each sensor (Sdata) and the basic parameters configuration are
presented in Table 1, it has been observed that the performance of SMA algorithm the minimum number of
sensor nodes that it has to be deployed.
Table 1. Table of parameters
Network Scale and Source Specification
Network Size 300 m * 300 m
Maximum Transmission Range 15 m
Topology Configuration Randomized
Total Sensor Node Number 600
Number of Target Sources ( nsource ) Default : 4
Data Rate at MAC layer Default : 7
Sensed Traffic Specification
Size of Sensed Data (Sdata ) Default: 4Kbytes
Sensed Data Interval 1 s
Task Duration (T task ) Default: 100 minutes
MA Specification
Fusion Factor (ρ ) 1
Reduction Ratio (r ) Default: 10%
Local Processing Time at Target Source (T proc ) Default: 50ms
Size of Processing Code ( S proc ) Default: 0.4Kbytes
4.4.1 Effect of the Number of Target Sources
In these experiments, we change the number of sources from 1 to 14 and keep all the other parameters in Table 1
unchanged. As shown in Figure 7, SMA algorithm consume less energy compared with MAWSN and CSWSN
algorithms, with maximum energy consumption equals to 1, 14 compared with 3,5 for MAWSN and 7,95 for
CSWSN .In Fig. 8, the average tcs increases relatively to nsources, however even though the performance of SMA
algorithm overcomes both CSWSN and MAWSN, the end to end delay introduces much more latency when the
number of source nodes becomes higher than 11 source nodes, and this is due to the additional time required by
the mobile agent to process data locally but it depends also on the maximum time that would take a mobile agent
to circulate all the source nodes in order to collect data. However in Figure 9, it can be observed clearly that the
cost (energy * delay) is the lowest compared with the other algorithms.

Vol.4, No.5, 2013
65
Figure 7. Comparison of total energy consumption
Figure 8. Comparison of the Average End-to-End delay
Figure 9. Comparison of the energy * delay

Vol.4, No.5, 2013
66
4.4.2 Effect of Size of Sensed Data
In these experiments, we change the size of the sensed data at each sensor (S data) from 0.5 KB to 4 KB in
0.5 KB interval. Figure10 shows that the energy consumption of SMA is always lower than that of
MAWSN and CSWSN when the size of data packet varied with maximum consumption of 0.24 (mWsec)
for SMA algorithm, 1.2 (mWsec) for MAWSN and 4.1 (mWsec) for CSWSN. Figure 11 shows that SMA
algorithm always outperforms CSWSN and MAWSN algorithm in terms of energy delay with preference
given to CSWSM in the first 0.5 KB interval, so the larger the data size is, the more efficient is the SMA
algorithm and once again in Figure 12, the cost (energy*delay) of SMA algorithm is much better compared
with MAWSN and CSWSN
Figure 10. Comparison of the Average End-to-End delay
Figure 11. Comparison of the energy * delay

Vol.4, No.5, 2013
67
Figure 12. Comparison of energy consumption
4.4.3 Effect of Reduction Ration
In the experiments, we change the reduction ration r from 0.1 to 0.4 and keep the other parameters in Table 1
unchanged. As observed in Figure 13, the energy consumed by SMA is much lower than CSWSN and MAWSN
and the same behavior is noted in Figure 14 where the end-to-end delay presents the minimum values compared
with the other algorithms Figure 15 we can clearly observe that when the Reduction Ratio is taking different
values, the performance of the SMA algorithm in terms of the cost(energy * delay) is still the best.
Figure13 comparison of energy consumption

Vol.4, No.5, 2013
68
Figure 14 comparison of the Average End-to-End delay
Figure 15 comparison of energy * delay
5. Conclusion and Perspectives
Sensor Networks are emerging as a great aid in improving the way of accessing to data. The previous studies
show that solar energy is an efficient alternate energy source for WSNs. The solar energy based routing concepts
increase the performance and life of WSNs compared with other conventional routing algorithms. In this paper,
we present the SMA routing algorithm for WSNs based on solar powered nodes, and as a future work, we would
be working on resolving the issue where the location of a sensor is changed or its battery is running out of energy.
SMA new tendency can be described as follows: every time the location of a sensor is changed, the sink Node
would send a beacon packet to alert the solar nodes. As soon as received, the solar nodes will drop the paths and
stop generating mobile agents to collect data until the new paths are received. In cases where a sensor node’s
battery power is dead, solar nodes will wait for the mobile agent until the timeout T is expired. T corresponds the

Vol.4, No.5, 2013
69
delay taken by a mobile agent to circulate the longest path towards the destination (solar node or sink
node) .Thus, when T is expired, solar node will send a beacon packet to the Sink Node in order to recalculate the
new paths. The development of this new methodology is still in progress.
.
References
Brussel, H., Wyns, J., Valckenaers, P., Bongaerts, L. & Peters, P. (1998), “Reference Architecture for Holonic
Manufacturing Systems: PROSA”, Computers in Industry 37(3), 255-274.
Chiara Buratti,Andria Conti,Davide Dardari & Roberto Verdone.(2009), “An Overview on Wireless Sensor
Networks Technology andEvolution”,Sensor 2009, 9, 6869-6896; doi:10.3390/s90906869,31 August.
Luis J. García Villalba, Ana L. Sandoval Orozco, Alicia Triviño Cabrera, and Cláudia J. Barenco Abbas.(2009),
“Routing Protocols in Wireless Sensor Networks”, State-of-the-Art Sensors Technology in Spain, Sensors 2009,
9(11), 8399-8421; doi:10.3390/s91108399.
Ankita Joshi , Lakshmi Priya.(2007), “A Survey of Hierarchical Routing Protocols in Wireless Sensor
Network,” International Journal of Engineering and Innovative Technology (IJEIT) ,Volume 1, Issue 5, May
2012.R. E. Sorace, V. S. Reinhardt, and S. A. Vaughn, “High-speed digital-to-RF converter,” U.S. Patent 5 668
842, Sept. 16, 2007.
Min Chen, Taekyoung Kwon, Yong Yuan, and Victor C.M.(2006), “Mobile Agent Based Wireless Sensor
Networks”, Journal of computers , vol. 1, NO. 1.M. Shell. (2002) IEEEtran homepage on CTAN.
Corke,P.Valencia,P.Sikka,T.Wark & L.Overs,Long-Duration.(2010), “solar-powered wireless sensor
networks”,Proceedings of the 4th Workshop on Embedded Networked Sensors. June 25-26,ACM,New
York,USA,pp:33-37FLEX Chip Signal Processor (MC68175/D), Motorola.
Min Chen, Wei Cai, Sergio González-Valenzuela, Victor C. M.(2009), “Leung,Balanced Itinerary Planning for
Multiple Mobile Agents in Wireless Sensor Networks”, Second International Conference, ADHOCNETS,
Victoria, BC, Canada,2010.A. Karnik, “Performance of TCP congestion control with rate feedback: TCP/ABR
and rate adaptive TCP/IP,” M. Eng. thesis, Indian Institute of Science, Bangalore, India.
Xiaoming Fan,W.Shaw and I Lee.(2007), “Layered clustering for solar powered wireless visual sensor
networks”,Proceedings of the 9th IEEE International Symposium on Multimedia,Dec.10-12,IEEE Computer
Society,Washington,DC,USA,PP:237-244.
Jean Breckling, Ed.(2008), “The Analysis of Directional Time Series: Applications to Wind Speed and
Direction”, ser. Lecture Notes in Statistics. Berlin, Germany: Springer, 2008, vol. 61.
Wei Cai ,M.Chen ,T.Hara,L.Shu , T.Kwon.(2011), “A genetic algorithm approach to multi-agent itinerary
planning in Wireless SensorNetworks”,Mobile Networks and Applications-Special issue on Wireless and
Personal Communications.
T.Voigt,H.Ritter,J.Schiller.(2003), “Utilizing Solar Power in Wireless Sensor Networks”, Universität Berlin,
Germany,Institut für Informatik, Local Computer Networks LCN '03. Proceedings. 28th Annual IEEE
International Conference on 20-21 Oct.2003,page:416- 422.
M.Chen,S.Gonzalez,V.C.M.Leung.(2007), “Applications and design issues of mobile agents in wireless
sensor networks”,IEEE Wireless Communication,vol 14,no.6,pp.20-26.

Energy efficient routing algorithm in wireless sensor networks

Recommended

More Related Content

What's hot (20)

Viewers also liked (11)

Similar to Energy efficient routing algorithm in wireless sensor networks (20)

More from Alexander Decker (20)

Recently uploaded (20)

Energy efficient routing algorithm in wireless sensor networks