20320130406025

Document Sample
20320130406025 Powered By Docstoc
					 INTERNATIONAL JOURNAL OF ADVANCED and Technology (IJARET), ISSN 0976 –
International Journal of Advanced Research in Engineering RESEARCH IN ENGINEERING
                              AND Volume 4, Issue 7, November – December (2013), © IAEME
6480(Print), ISSN 0976 – 6499(Online)TECHNOLOGY (IJARET)


ISSN 0976 - 6480 (Print)
ISSN 0976 - 6499 (Online)
Volume 4, Issue 7, November - December 2013, pp. 207-215
                                                                           IJARET
© IAEME: www.iaeme.com/ijaret.asp
Journal Impact Factor (2013): 5.8376 (Calculated by GISI)                  ©IAEME
www.jifactor.com




   A NEW ENHANCED LEACH ROUTING PROTOCOL FOR WIRELESS
 SENSOR NETWORK BASED ON GAUSSIAN DISTRIBUTION OF RANDOM
                        NUMBERS

                                 Ankit Thakkar1, Ketan Kotecha2
         1
           Assistant Professor, CSE Department, Institute of Technology, Nirma University,
                                  Ahmedabad - 382 481, Gujarat, India
     2
       Director, Institute of Technology, Nirma University, Ahmedabad - 382 481, Gujarat, India



ABSTRACT

        Wireless Sensor Network (WSN) is used to monitor hostile environments such as military
applications, habitat monitoring etc., where it is difficult to replace or recharge the batteries of the
sensor nodes once they deployed. Thus, prolonging network lifetime of the WSN is the biggest
challenge for the researchers across the globe. Clustered based routing protocol such as Low Energy
Clustering Hierarchy (LEACH), provides network longevity. We have proposed a new routing
protocol based on the Gaussian distribution of the random numbers, that is one of the parameter used
to elect the cluster heads. Extensive simulations are carried out to verify the validity of the proposed
approach. Simulation results show that our proposed approach is energy efficient compared to
LEACH and it sustain with the varying node density.

Key words: LEACH, Energy Efficient Routing, Cluster Head Selection, Wireless Sensor Network,
Gaussian Distribution, Random Number.

1. INTRODUCTION

       Wireless Sensor Network (WSN) becomes popular to provide cost effective solutions because
of enhancement in the sensor and communication technology. These cost effective solutions attract
researchers to provide solutions to number of civilian and military applications. Sensor Nodes
monitor specific parameter from the Region of Interest (ROI) and pass this information to the Sink or
Base Station (BS).
       Sensor Nodes are battery operated; and have low communication and computational capacity.
These nodes remain unattained once they are deployed in the hostile environments. Also, cost of

                                                  207
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME

battery replacement of the sensor nodes are very expensive (Niculescu, D. (2005)). Hence, network
lifetime enhancement is a challenging issue for the researchers across the globe. It has been proven
that energy required for communication is very high compared to computation (Akyildiz, I. F., Su,
W., Sankarasubramaniam, Y., & Cayirci, E. (2002)). Hence, communication between the sensor
nodes should be minimized to enhance the network lifetime. Also, energy required for
communication is directly proportionate to the square or quad of the distance between the sender and
desired recipient.
       Low Energy Clustering Hierarchy (LEACH) (Heinzelman, W. R., Chandrakasan, A., &
Balakrishnan, H. (2000, January)), is one of the cluster based routing protocol that provide
prolonged network lifetime. This protocol may be used as a benchmark to compare any new cluster
based routing algorithm. In this paper, we have proposed a new clustered routing protocol based on
Gaussian distribution of random numbers that is used to elect cluster head. As per our knowledge, we
are the first to make an attempt to elect cluster head using Gaussian distribution of random numbers.
       Rest of the paper is organized as follows: Related Work is described in Section 2; System and
Energy models are given in Section 3; Proposed approach is discussed in Section 4; Simulation
Strategy and Result Discussion are presented in Section 5 and Concluding remarks are given in
Section 6.

2. RELATED WORK




Figure 1: Phases of LEACH protocol (Megerian, S., & Potkonjak, M. (2003), Yassein, M. B., A. Al-
                        zou'bi, Khamayseh, Y., & Mardini, W. (2009))

        In LEACH (Heinzelman, W. R., Chandrakasan, A., & Balakrishnan, H. (2000,
January)), authors have proposed energy efficient clustered based routing scheme, in which, role of
cluster head (CH) is rotated between nodes to achieve uniform energy depletion through load
balancing. Each sensor node elects itself as a CH with certain probability. This status is informed by
the CH nodes to the other nodes in the network. After getting status message from the CH nodes,
non-CH nodes select one of the CHs as its own cluster head for which minimum energy is required
for communication. CH nodes receive data from the non-CH nodes, fuse it and send to the Base
Station. At the commencement of the each round, a node selects a random number which is
uniformly distributed pseudo random number between 0 and 1. If random number is less than
threshold value T(n) as given by equation 1, then node is elected as a CH for the current round.
LEACH ensures that each node becomes CH only once every 1/p rounds, where p is the desired
percentage of cluster heads known a priori to the algorithm. Phases of LEACH protocol is show in
Figure 1.



                                                 208
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME



                                                                               (1)



        In CVLEACH (Thakkar, A., & Kotecha, K. (2012)), authors have proposed energy efficient
cluster head election scheme using overhearing property of the sensor nodes. In ALEACH (Ali, M.
S., Dey, T., & Biswas, R. (2008, December)), authors have modified threshold value T(n) which is
given by Equation 2. Like LEACH, ALEACH also works in rounds. Each round begins with Cluster
Setup phase. During cluster setup phase, a node selects a random number between 0 and 1. If
selected random number is less than threshold value T(n), then node will declare itself as a cluster
head where T(n) is given by equation 2.


                                                                               (2)

where Gp and CSp is given by equations 3 and 4 respectively. Gp and CSp refer to general
probability and current state probability.


                                                                               (3)


where k/N refers to the desired percentage of cluster heads during each round and Ecurrent and Emax is
remaining energy and maximum energy of a node respectively.

                                                                               (4)


        In WALEACH (Thakkar, A., & Kotecha, K. (2012)), authors have modified threshold
value T(n) by assigning weight (importance) to Gp and CSp. In WCVALEACH (Thakkar, A., &
Kotecha, K. (2012)), authors have improved by WALEACH by assigning importance to the
parameters used to calculate threshold along with taking advantage of over hearing property of the
sensor nodes. In EDACH (Kim, K. T., & Youn, H. Y. (2005, January)), authors have modified
Equation 1 to calculate threshold value to elect the cluster head. They have assigned different values
to p, depending upon node’s distance from the BS i.e. near, medium or far. In (Handy, M. J., Haase,
M., & Timmermann, D. (2002)), authors have modified T(n) in Equation 1 to elect cluster head by
considering ratio of node’s current energy with respect to initial energy. In REEH (Sehgal, L., &
Choudhary, V. (2011)), authors have modified threshold value T(n) which is given by Equation 5.


                                                                               (5)


       In (Loscri, V., Morabito, G., & Marano, S. (2005, September)), authors have proposed
two-level clustering hierarchy, in which CH nodes collect data from member nodes as in LEACH,
but CH nodes do not transmit fused data directly to BS; but transmit through other CH to save the
energy of nodes. In (Xiangning, F., & Yulin, S. (2007, October)), authors have proposed energy

                                                 209
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME

based LEACH protocol, in which all nodes are having equal probability to become a CH for the first
round. After first round, all nodes are having different energy levels. Nodes are having higher energy
after first rounds are having more probability to be a CH in the next round. In (Guifeng, W., Yong,
W., & Xiaoling, T. (2009, October)), authors have calculated T(n) value by considering current
energy of a node and initial energy of a node. After election of a cluster head, authors have used
ACO approach to transmit data from CH to BS. In (Lee, S., Yoo, J., & Chung, T. (2004,
November)), authors have proposed a new distributed clustering and data aggregation algorithm
CODA based on distance from the sink in the wireless sensor network. A more detailed survey on
variations of LEACH algorithm can be found at (Kumar, V., Jain, S., & Tiwari, S. (2011)).
         LEACH and its variations presented in the paper, have tried to prolong the network lifetime
by formulating new bounds for threshold T(n) by considering different parameters such as energy
and distance. In these algorithms, a node will generate a random number between 0 and 1; if the
random number is less than T(n) then that node elects itself as a CH for the current round. Each and
every algorithm presented in the paper have tried to modified T(n), and compared it with a random
number that is based on uniform random number generation scheme. In this paper, we have proposed
our protocol to elect CH using Gaussian random number.

3. PREREQUISITE

3.1 System Model
    In this paper, we have considered a sensor network of N nodes that area uniformly distributed in
the region of MxM m2. Following assumptions are made about the sensor nodes and underlying
network (Chen, G., Li, C., Ye, M., & Wu, J. (2009)).

   1. Base Station (BS) is located in the center of the node deployment area and it is having infinite
      amount of energy.
   2. BS and sensor nodes are stationery once they are deployed in the Region of Interest (ROI).
   3. All sensor nodes are homogeneous and are assigned unique identifier.
   4. All sensor nodes are having limited amount of energy.
   5. All sensor nodes are capable of transmitting with different power levels depending upon the
      distance from the desired recipient.
   6. Communication links are symmetric.
   7. Cluster Head always receive highly correlated data from the member nodes. Thus, data
      aggregation is possible.

3.2 Energy Model
       We have considered first order of radio model as given in (Heinzelman, W. B.,
Chandrakasan, A. P., & Balakrishnan, H. (2002), Heinzelman, W. R., Chandrakasan, A., &
Balakrishnan, H. (2000, January), Chen, G., Li, C., Ye, M., & Wu, J. (2009)) for communication
energy expenditure. We have used free space model (d2 power loss) and multipath fading model (d4
power loss) where d is the distance between the transmitting and receiving nodes. The energy
expenditure to transmit l-bit data packet over distance d is given by equation 6.


                                                                               (6)




                                                 210
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME

       The electronics energy, Eelec, depends on factors such as the digital coding, and modulation,
whereas the amplifier energy, l∈fsd2 or l∈fsd4, depends on the transmission distance and the
acceptable bit-error rate (Chen, G., Li, C., Ye, M., & Wu, J. (2009)). To receive this message, the
radio expends energy as given by equation 7. We have also assumed that a cluster head consumes
EDA (nJ/bit/signal) amount of energy for data aggregation.

                                                                                (7)

4. PROPOSED APPROACH

        Like LEACH, our proposed approach also works in round. Round is the duration for which a
particular node works as a CH. Round consists of two phases. (a) Cluster Head election phase and
(b) Steady State Phase. Both of these phases are described in the following subsections.

4.1 Cluster Head Election Phase
       At the beginning of each round, a node selects a random number between 0 and 1. Unlike
LEACH, this random number is generated using Gaussian distribution with mean 0 and variance 1.
Random number must be between 0 and 1; otherwise new random number is selected by the node.
This process is continued until the node found a random number between 0 and 1. Once a random
number is generated between 0 and 1 for a particular node, it is compared with threshold T(n) which
is given by Equation 1, in that r is the current round number; p is the desired percentage of the CHs
and G is the set of nodes those had not been cluster heads since last r rounds. A node elects itself as
a CH, if the random number less than T(n), otherwise it works as a member node (non-CH node) for
the current round. CH nodes inform their status to other nodes within the network. Non-CH node
selects one of the CH nodes for which minimum communication energy expenditure is required; and
sends a join message to the selected CH. After getting join messages from the Non-CH nodes, CH
nodes prepares TDMA schedule and inform to the member nodes of their own cluster.

4.2. Steady State Phase
       During the steady state phase, all non-CH nodes turn off their radio to save their energy,
except the time slot for which they have to transmit data to CH. CHs perform data aggregation and
send aggregated data to BS after receiving data from the member nodes. This data aggregation
approach helps to save energy of the CH because less number of bits transmitted by CHs. At the end
of Steady State phase a new round begins with CH election phase.
       There is a high energy drains for the cluster heads because they have to do CH announcement
to the other nodes, reception of CH Join message, TDMA schedule announcement to the nodes who
have sent CH Join messages, data reception from the member nodes and finally data transmission to
BS after performing data aggregation. Thus, to enhance the network life time, role of the CH is
rotated between all active nodes of the network. Thus, set of nodes xj who are the cluster head for the
round j is different than the set of nodes xj-1 who had been cluster head for the previous round.

5. SIMULATION STRATEGY AND RESULT DISCUSSION

       We have simulated our proposed approach – the Gaussian LEACH and compared it with the
LEACH protocol for a random network of 100 nodes, 200 nodes and 500 nodes deployed in a region
of 100m x 100m. We have also monitor the effect of percentage of total nodes those are cluster heads
with p=0.05, p=0.10 and p=0.20. Since, both the protocols are stochastic in nature; results of two
successive runs will not be same. Hence, we have simulated both the protocols five times for each
                                                 211
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME

network configuration and mean value of the interested parameter is plotted in the result. Simulation
parameters are given in the Table 1.
        To compare Gaussian LEACH with LEACH, we have recorded time (in rounds) period for
the death of first node, death of 20% nodes, death of 40% nodes, death of 60% nodes and death of
100% nodes with p=0.05, p=0.10 and p=0.20 and node density of 100, 200 and 500 nodes. These
results are shown in Figure 2, 3 and 4 for the network of 100 nodes, 200 nodes and 500 noes
respectively.
        LEACH outperforms Gaussian LEACH for the First Node Dies (FND) i.e. death of first node
after network deployment; while Gaussian LEACH outperforms LEACH for the death of 20% nodes,
40% nodes, 60% nodes, 80% nodes and 100% nodes for a network of 100, 200 and 500 nodes with
percentage of cluster heads are 5%, 10% and 20%.

                                  Table 1. Simulation Parameters
                        Parameter Name                                Value
                         Simulation Area              100m x 100m
                           Total Nodes                    (1) 100
                                                          (2) 200
                                                          (3) 500
                          Cluster Heads                   (1) 5%
                   ( percentage of total nodes)           (2) 10%
                                                          (3) 20%
                          Initial Energy              0.5 J/Node
               Transmitter Electronics and Receiver 50 nJ/bit
                           Electronics
                       Transmit Amplifier             100 pJ/bit/m2
                  Energy for Data Aggregation         5nJ/bit/message
                          Message Size                2000-bit
                         Simulation runs              Each network configuration was run 5
                                                      times




                       Figure 2. Death of Nodes for a network of 100 nodes


                                                212
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME




                       Figure 3. Death of Nodes for a network of 200 nodes




                       Figure 4. Death of Nodes for a network of 500 nodes

        The longevity of network lifetime varies for different percentage of cluster heads and
different node density as both protocols are stochastic in nature. From the results, we can say that
Gaussian LEACH enhances network lifetime compared to LEACH.

                                                213
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME

6. CONCLUSION

        In this paper, we have proposed Gaussian LEACH protocol, which is distributed in nature,
sustain with the node density and provides improved network lifetime compared to LEACH protocol.
In Gaussian LEACH, we have generated random numbers between 0 and 1 using Gaussian
distribution with mean 0 and variance 1. These numbers are compared with the threshold T(n) which
is same as LEACH to elect cluster heads. Our protocol enhances overall network lifetime compared
to LEACH but First Node Dies (FND) is not as good as LEACH. In future, we will propose the
solution to improve FND parameter for the Gaussian LEACH.

REFERENCES

 1.    Niculescu, D. (2005). Communication paradigms for sensor networks. Communications
       Magazine, IEEE, 43(3), 116-122.
 2.    Akyildiz, I. F., Su, W., Sankarasubramaniam, Y., & Cayirci, E. (2002). A survey on sensor
       networks. Communications magazine, IEEE, 40(8), 102-114.
 3.    Heinzelman, W. R., Chandrakasan, A., & Balakrishnan, H. (2000, January). Energy-efficient
       communication protocol for wireless microsensor networks. In System Sciences, 2000.
       Proceedings of the 33rd Annual Hawaii International Conference on (pp. 10-pp). IEEE.
 4.    Megerian, S., & Potkonjak, M. (2003). Wireless sensor networks. Encyclopedia of
       Telecommunications.
 5.    Yassein, M. B., A. Al-zou'bi, Khamayseh, Y., & Mardini, W. (2009). Improvement on
       LEACH Protocol of Wireless Sensor Network (VLEACH).JDCTA, 3(2), 132-136.
 6.    Thakkar, A., & Kotecha, K. (2012). CVLEACH: Coverage based energy efficient LEACH
       Algorithm. International Journal of Computer Science and Network (IJCSN) Volume, 1.
 7.    Ali, M. S., Dey, T., & Biswas, R. (2008, December). ALEACH: Advanced LEACH routing
       protocol for wireless microsensor networks. In Electrical and Computer Engineering, 2008.
       ICECE 2008. International Conference on (pp. 909-914). IEEE.
 8.    Thakkar, A., & Kotecha, K. (2012). Computer Science & Information Technology
       (CS & IT). WALEACH: Weight Based Energy Efficient Advanced LEACH Algorithm, 2 (4),
       117-130.
 9.    Thakkar, A., & Kotecha, K. (2012). WCVALEACH: Weight and Coverage Based Energy
       Efficient Advanced LEACH Algorithm. Computer Science & Engineering, 2(6).
 10.   Kim, K. T., & Youn, H. Y. (2005, January). Energy-driven adaptive clustering hierarchy
       (EDACH) for wireless sensor networks. In Embedded and Ubiquitous Computing–EUC 2005
       Workshops (pp. 1098-1107). Springer Berlin Heidelberg.
 11.   Handy, M. J., Haase, M., & Timmermann, D. (2002). Low energy adaptive clustering
       hierarchy with deterministic cluster-head selection. In Mobile and Wireless Communications
       Network, 2002. 4th International Workshop on (pp. 368-372). IEEE.
 12.   Sehgal, L., & Choudhary, V. (2011). REEH: Residual Energy Efficient Heterogeneous
       Clustered Hierarchy Protocol for Wireless Sensor Networks. International Journal of Scientific
       & Engineering Research, 2(12), 1-5.
 13.   Guifeng, W., Yong, W., & Xiaoling, T. (2009, October). An ant colony clustering routing
       algorithm for wireless sensor networks. In Genetic and Evolutionary Computing, 2009.
       WGEC'09. 3rd International Conference on (pp. 670-673). IEEE.
 14.   Loscri, V., Morabito, G., & Marano, S. (2005, September). A two-levels hierarchy for low-
       energy adaptive clustering hierarchy (TL-LEACH). In IEEE Vehicular Technology
       Conference (Vol. 62, No. 3, p. 1809). IEEE; 1999.

                                                 214
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 7, November – December (2013), © IAEME

 15. Xiangning, F., & Yulin, S. (2007, October). Improvement on LEACH protocol of wireless
     sensor network. In Sensor Technologies and Applications, 2007. SensorComm 2007.
     International Conference on (pp. 260-264). IEEE.
 16. Kumar, V., Jain, S., & Tiwari, S. (2011). Energy efficient clustering algorithms in wireless
     sensor networks: A survey. IJCSI International Journal of Computer Science Issues, 8(5),
     1694-0814.
 17. Lee, S., Yoo, J., & Chung, T. (2004, November). Distance-based energy efficient clustering
     for wireless sensor networks. In Local Computer Networks, 2004. 29th Annual IEEE
     International Conference on (pp. 567-568). IEEE.
 18. Chen, G., Li, C., Ye, M., & Wu, J. (2009). An unequal cluster-based routing protocol in
     wireless sensor networks. Wireless Networks, 15(2), 193-207.
 19. Heinzelman, W. B., Chandrakasan, A. P., & Balakrishnan, H. (2002). An application-specific
     protocol architecture for wireless microsensor networks. Wireless Communications, IEEE
     Transactions on, 1(4), 660-670.
 20. Neeraj Tiwari, Rahul Anshumali and Prabal Pratap Singh, “Wireless Sensor Networks:
     Limitation, Layerwise Security Threats, Intruder Detection”, International Journal of
     Electronics and Communication Engineering & Technology (IJECET), Volume 3, Issue 2,
     2012, pp. 22 - 31, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472.
 21. Yogesh V Patil, Pratik Gite and Sanjay Thakur, “Automatic Cluster Formation and Assigning
     Address for Wireless Sensor Network”, International Journal of Computer Engineering &
     Technology (IJCET), Volume 4, Issue 4, 2013, pp. 116 - 121, ISSN Print: 0976 – 6367,
     ISSN Online: 0976 – 6375.
 22. Rashid M. Awadi, Rawya Y. Rizk, Mohamed I. Habib and Amira A. M. Elsonbaty, “An
     Efficient Cluster Head Selection Scheme for Dynamic Sleep Time in Wireless Sensor
     Network”, International Journal of Electronics and Communication Engineering &
     Technology (IJECET), Volume 4, Issue 6, 2013, pp. 1 - 13, ISSN Print: 0976- 6464,
     ISSN Online: 0976 –6472.
 23. Mohanaradhya, Sumithra Devi K A and Andhe Dharani, “Distance Based Cluster Head
     Section in Sensor Networks for Efficient Energy Utilization”, International Journal of
     Advanced Research in Engineering & Technology (IJARET), Volume 4, Issue 1, 2013,
     pp. 50 - 58, ISSN Print: 0976-6480, ISSN Online: 0976-6499.
 24. Preetee K. Karmore, Supriya S. Thombre and Mr. Gaurishankar L. Girhe, “Review on
     Operating Systems and Routing Protocols for Wireless Sensor Networks”, International
     Journal of Computer Engineering & Technology (IJCET), Volume 4, Issue 3, 2013,
     pp. 331 - 339, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.




                                              215

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:12/31/2013
language:English
pages:9