Simulation and Research on Data Fusion Algorithm of the Wireless Sensor Network Based on Ns2
In:
Submitted By aneesh91 Words 3532 Pages 15
2009 World Congress on Computer Science and Information Engineering
Simulation and Research on Data Fusion Algorithm of the Wireless Sensor
Network Based on NS2
Junguo Zhang, Wenbin Li, Xueliang Zhao, Xiaodong Bai, Chen Chen
Beijing Forestry University, 35 Qinghua East Road, Haidian District,Beijing, 100083 P.R.China information which processed by the embedded system to the user terminals by means of random selforganization wireless communication network through multi-hop relay. Thus it authentically achieves the purpose of ‘monitor anywhere and anytime’. The basic function of sensor network is gathering and sending back the information of the monitoring areas which the relevant sensor nodes are set in. But the sensor network node resources are very limited, which mainly embodies in battery capacity, processing ability, storage capacity, communication bandwidth and so on.
Because of the limited monitoring range and reliability of each sensor, we have to make the monitoring areas of the sensor nodes overlapped when they are placed in order to enhance the robustness and accuracy of the information gathered by the entire network. In this case, certain redundancy in the gathered data will be inevitable. On the way of sending monitoring data by multi-hop relay to the sink nodes (or base stations) which are responsible to gather the data. It is necessary to reduce the redundant information by fusion processing. Data fusion is generally defined as a process that lots of data gathered from multiple sources or information are processed and combined in order to achieve more efficient data and ones that can better meet the requirements of the users. In most applications of wireless sensor network, sometimes we only concern about the monitoring results rather than gathering a lot of original data. Data fusion technology is an efficient method to solve this kind of problems.
Abstract
The Wireless Sensor Network technology has been used widely; however the limited energy resource is one of the bottlenecks for its application. To enhance the robustness and accuracy of the information which is obtained by the entire network, certain redundancy exists in the data collected from the sensor node and thus data fusion processing is needed to reduce the redundant information. Firstly the function and categories of data fusion were introduced in this paper, and also the relation between data fusion and protocol layers of the sensor network were described.
Meanwhile various data fusions which combine the route patterns in the network, especially dynamic cluster organization algorithm were emphasized on.
On the basis of analyzing the principle of the LEACH protocol, a recent typical data fusion algorithm, the simulation and contrastive analysis of the LEACH protocol and its ameliorated algorithm, the LEACH-C protocol, were implemented by using the NS2 network simulation software. The simulation results suggest that the LEACH-C protocol can effectively solve the problems such as the uncertainty of the number of the cluster header in the LEACH algorithm, lack of consideration of the energy state of the nodes in the construction stage of dynamic cluster and so on.
1
1. Introduction
2. Wireless sensor network data fusion technology Wireless sensor network consists of a large number of various mini integrated sensor nodes distributing in the monitoring area, which can be real-time perceiving and monitoring all kinds of information about the environment or objects. They send the perceived
In the research field of wireless sensor network at present, many scholars relate data fusion technology to the study of protocol hierarchy. Take the design of application layer for example, in order to achieve the fusion effect distributed database technology is used to screen the collected data step by step; in the study of network layer, a lot of routing protocols use data
1. Corresponding Author: Wenbin Li, Leewb@bjfu.edu.cn.
2. This work was supported in part by Import Project under China
State Forestry Administration (2008-4-53), Forestry Special Public
Sector Research Funds Project(200804029), and National Student
Innovation Plan Project.
Authorized licensed use limited to: Mississippi State University. Downloaded on November 27, 2009 at 15:22 from IEEE Xplore. Restrictions apply.
data to the CH nodes of the clusters which they are in.
For reducing the transmission of the redundant data, the CH nodes send the data to sink nodes after data fusion processing. After a certain time interval, another set of nodes are selected as new CH nodes randomly.
Thus each non-CH node only needs to know the CH’s information of the cluster which it is in, and the CH also only needs to maintain a small routing table.
Figure 1 shows a typical layered structure of WSN using the LEACH algorithm. In practical application, we can also construct more layers according to our needs. fusion mechanism to reduce the amount of data transferred[1-4]; in the data link layer, the sending conflicts and head overhead in the MAC layer are reduced combined of MAC protocol to save energy.
In the structure of wireless sensor network system, routing technology of the network layer has a significant influence on the performance of WSN.
With the research development of WSN at home and abroad, many routing protocols have been proposed.
From the perspective of network topology we can generally divide them into two categories: the plane routing protocol and the hierarchical routing protocol.
Typical plane routing algorithms include Flooding[5],
DD (Directed Diffusion), SAR (Sequential Assignment
Routing), SPIN (Sensor Protocols For Information Via
Negotiation), Rumor Routing and so on. The most serious disadvantage of the plane routing lies in: there are no manage nodes in the network, lacking of optimal management of the communication resources, complicated algorithms for self-organizing and cooperative work, slow response to the dynamic changes of the network etc. The hierarchical routing protocols mainly include LEACH (Low Energy
Adaptive Clustering Hierarchy)[2], TEEN (Threshold
Sensitive Energy Efficient Sensor Network Protocol)[3],
APTEEN (Adaptive Periodic Threshold-Sensitive
Energy Efficient Sensor Network Protocol) and so on.
The hierarchical routing protocols can efficiently decrease the cost of maintaining the routing maintenance system, improve the scalability of the network and the utilization rate of the channel space, and has a significant application value on the application system of large-scale network.
The LEACH protocol is based on the concept of hierarchical routing and is the first cluster-based routing protocol proposed in WSN. Most of the following hierarchical routing protocols, such as
TEEN, PEGASIS etc., are all developed on the basis of
LEACH protocol. Thus it is helpful for getting further cognition of other hierarchical routing algorithms to have research on the LEACH algorithm. So it has a good typicality and representative to choose the
LEACH protocol as the research object[6].
Figure 1. Hierarchical structure of LEACH
In the LEACH algorithm, each node has to be the
CH alternately for the sake of avoiding the energy of
CH being consumed too fast. Thus the implementation of this algorithm is separated into several rounds. Each round also can be divided into a construction stage and a stable transmission stage. The construction stage is a cluster construction stage, while in the stable transmission stage each node sends the data to the CH.
In the construction stage, it is random to choose a node as the CH node, of which the randomness ensures that the high cost of data transmission between the CH and the sink node is evenly allocated to all the sensor nodes.
The CH node broadcasts the information to the surrounding, while other nodes will decide to join in a cluster according to the intensity of the information they have received, and inform the corresponding CH nodes. In the stable transmission stage, nodes continuously collect monitoring data and send them to the CH. Then the data will be sent to the sink node by
CH after its necessary fusion processing.
The time flowchart is shown as Figure 2. After the stable stage sustains for a period, the network moves forward into the next working round and reselect CH.
For reducing the extra energy cost produced by dividing clusters, the stable stage is much longer than the construction stage.
2.1. The LEACH routing protocol algorithm
The realization of LEACH algorithm is mainly by selecting cluster-head nodes randomly and then to share relay communication service on average. In the
LEACH algorithm, the nodes are self-organized into different clusters, with electing Cluster Header (CH) nodes respectively and average the energy consumption by rotating the CH nodes. Each cluster can only have one CH. All non-CH nodes send the
Set-up
Steady-state
Frame
Round
Time
Figure 2. Time line showing LEACH operation
67
Authorized licensed use limited to: Mississippi State University. Downloaded on November 27, 2009 at 15:22 from IEEE Xplore. Restrictions apply.
the LEACH algorithm in the stable data transmission stage. Comparing to the general plane routing protocol and the static clustering algorithm, the LEACH protocol has a lot of advantages such as the layered structure, local data fusion processing, dynamic CH allocation and so on. Especially when dealing with those data which have high correlations among them, a large amount of redundant data will be eliminated because of data fusion, which makes LEACH have a better performance on energy consumption[7].
3. Simulation and research on data fusion algorithm of the wireless sensor network based on NS2
3.1. NS2 introduction
NS2 (Network Simulator version2) was developed by UC-Berkeley in United States. It is an open source and free software simulation platform which aims at network technology. Essentially, it’s a discrete event simulator. It has a virtual clock, and all the simulations are driven by discrete events. Researchers can conveniently carry out the development of network technology by using it. What’s more, it has contained quite rich modules and almost concerned all the aspects of network technology up to today[9]. Therefore,
NS2 has become the network simulation software which is now widely used in academic circles.
2.2. The LEACH-C routing protocol algorithm
LEACH-C (LEACH-centralized) protocol is a kind of centralized routing protocol which is proposed on the basis of the LEACH algorithm. In this protocol, the base station is in charge of choosing the CH according to the global information. Each node sends the information on its location and current energy state to the base station; then according to the information that provided by all the nodes, the base station calculates the average energy. Those nodes whose current energy is lower than the average energy cannot be the candidate CH. It’s a NP problem that choosing an appropriate quantity of cluster-headers with optimal locations from the remaining nodes to form a set.
Following the principal of minimal square sum of distance between all the nodes to the CH, the base station uses simulated annealing algorithm to solve the
NP problem[8].
The LEACH-C algorithm is a central control algorithm, which embodies on the clustering algorithm.
In the network construction stage of LEACH-C, each node sends the information on its present location and remaining energy to the base station. The base station is responsible for the CH nodes selection and clustering calculation. To make a better clustering, the base station has to make sure that all the nodes in the network have an even energy consumption. Therefore the base station calculate the average energy of all the nodes in the network, and those nodes whose remaining energy is lower than the average have no chances to be the CH nodes. Moreover, the base station can know the density of the nodes within certain area by the coordinates of the nodes. At last, the simulated annealing algorithm is run on the base station nodes to calculate the optimal clustering and
CH nodes. After the base station has determined the clustering and CH nodes, it will broadcast the IDs of the CH nodes and the clustering information through the network. On receiving the broadcast information from the base station, the node will turn into CH node if its ID is just the CH Node’s ID in the information, and other nodes will receive the IDs of their own CH nodes as well. The LEACH-C algorithm is the same as
3.2. Simulation and analysis on the LEACH protocol and the LEACH-C protocol based on
NS2
To valuate the performance of the protocol, this paper did simulation experiments on LEACH and
LEACH-C based on NS2. The experiments were carried on under the same energy module and the same quantity of sensor nodes, with a hundred of nodes distributed randomly in a 100m ⋅100m area where the abscissa x and the ordinate y are both from 0 to 100, and the position coordinates of the base station are
BS (50,175) as shown in Figure 3.
Figure 3. Positions of nodes and base station
In order to comprehensively appraise the effect of different protocols, the experiments compared LEACH with LEACH-C in network energy consumption,
68
Authorized licensed use limited to: Mississippi State University. Downloaded on November 27, 2009 at 15:22 from IEEE Xplore. Restrictions apply.
decide the selection of cluster headers and cluster construction. Thus less energy would be used to complete the data transmission. In addition, the simulated annealing algorithm used by the base station in the LEACH-C protocol ensures that the number of the cluster header (k) is always five in every dynamic-clustering round. Though the expectation of the cluster header (k) is five also in the
LEACH protocol, but it couldn’t always be that condition in the practical operation.
The time-survival nodes quantity and the data quantity received by the base station-survival nodes quantity schematic diagrams of the two protocols are shown as Figure 6 and Figure 7 respectively.
death rate of the node, data quantity received by the base station and quantitative changes of the cluster headers. To eliminate the experimental error caused by randomness, the experiment was repeated for ten times and the average was taken as the final result.
The schematic diagram of the time-data quantity received by the base station and another one of energy consumption-data quantity received by the base station are shown as Figure 4 and Figure 5 respectively. From the figures, we can see that the data quantity which the LEACH-C protocol received is more than that did by the LEACH protocol. It is obviously that either the data quantity that the base station received per unit time or the data quantity transferred to the base station per unit energy consumption, the LEACH-C protocol is both better than the LEACH protocol.
100
90
LEACH-C
80
4
Number of nodes alive
Number of data received at the base station
4.5
x 10
4
3.5
3
LEACH-C
2.5
LEACH
60
50
40
30
20
2
10
LEACH
1.5
0
1
0
100
200
300
Time(s)
400
500
100
200
300
Time(s)
400
500
600
600
100
Figure 4. Total amount of data received at the
BS station over time
LEACH-C
90
Number of nodes alive
80
4
4.5
0
Figure 6. Number of nodes alive over time
0.5
0
Number of data received at the base station
70
x 10
4
3.5
3
LEACH-C
LEACH
70
60
50
40
30
20
2.5
10
2
0
LEACH
1.5
1
0
20
40
60
80
100
120
Energy(J)
140
160
180
0.5
1
1.5
2
2.5
3
Number of data received at BS
3.5
4
4.5
4
x 10
Figure 7. Number of nodes alive per amount of data sent to the BS
From the figures we can see that the LEACH-C protocol is better than the LEACH protocol. Either in per unit operation time or data quantity per unit that the base station received, the survival nodes of the
LEACH-C protocol are always more than that of the
LEACH protocol in the network. If regard the death time of the first node in the network as a standard of
0.5
0
0
200
Figure 5. Total amount of data received at the
BS per given number of energy
In the LEACH-C protocol, the base station has the information of the locations and energy of all the nodes in the network, which makes it better to
69
Authorized licensed use limited to: Mississippi State University. Downloaded on November 27, 2009 at 15:22 from IEEE Xplore. Restrictions apply.
sensor network, many scholars have related the data fusion technology to the research of protocol layers, and the routing technology of the network layer has a significant influence on the performance of the wireless sensor network. The clustering routing technology which regards the LEACH protocol as typical has the advantages of convenient topology management, high efficiency of energy utilization and simple data fusion, thus it becomes the routing technology that is studied with an emphasis.
In this paper, the LEACH protocol and its ameliorated algorithm LEACH-C protocol were simulated and analyzed contrastively by using the NS2 network simulation software on the network energy consumption, death rate of the nodes, the amount of data received by the base station received and the change of the number of the cluster header. The simulation results suggest that the LEACH-C protocol can effectively solve the problems such as the uncertainty of the number of the cluster header, lack of consideration of the energy state of the nodes in the construction stage of dynamic cluster and so on in the
LEACH algorithm.
the life period of the network, both figure 6 and figure
7 suggest that the network life period of the
LEACH-C protocol is much longer than that of the
LEACH protocol. From the perspective of death rate of the network nodes, the LEACH-C protocol’s is also slower than the LEACH protocol’s. So the LEACH-C protocol can prolong the network life period efficiently.
The time-number of cluster header schematic diagram of the two protocols is shown as Figure 8.
14
Cluster-Head Number
12
10
LEACH
8
6
4
2
LEACH-C
0
50
100
150
200
250
300
Time(s)
350
400
450
500
Figure 8. Number of cluster head over time for
LEACH and LEACH-C
5. References
Because the LEACH protocol gets the number of the cluster header by comparing the random number with the changing threshold which relates to the number of rounds, the number can only be approximately equal to the number of optimal cluster
k. From the figure, the number of cluster header of the
LEACH protocol is about five (the optimal value of one hundred nodes), and change in the closed interval
[2, 13]. The centralized type used by the LEACH-C protocol ensures the optimal number of the cluster header can be always maintained at five. Therefore we can find out that the LEACH-C protocol makes sure the optimization of number of the cluster header, which optimizes the performance of the protocol.
[1] Intanagonwiwat C, Govindan R, and Estrin D.Directed
Diffusion: a scalable and robust communication paradigm for sensor networks[C].In: Proceedings of the ACM
MobiCom’00.Boston: ACM Press,2000.56-57
[2] Heinzelman W R, Chandrakasan A, Balakrishnan
H.Energy–Efficient Communication Protocol for Wireless
Microsensor Networks[C].In:Proc.33rd Hawaii Int’l Conf on
System Sciences(HICSS’00).Maui: IEEE Press,2000.1-10
[3] Manjeshwar A, Agrawal D P.TEEN:A routing protocol for enhanced efficiency in wireless sensor networks[C].In:Proc.15th Int’l Parallel and Distributed
Processing
Symp(IPDPS.01).San
Francisco:IEEE
Press,2001.4
[4] Gupta G, Younis M. Fault-tolerent clustering of wireless sensor networks[C].In:Proc IEEE Wireless Communications and Networking(WCNC 2003).Volume 3,16-20 March,2003
[5] Hedetniemi S,Liestman A. A suery of Gossiping and
Broadcasting
in
Communication
Networks[J],
Networks,1988,18(4):319-349
[6] XIE D M. The Study of LEACH Routing Protocol for
Wireless Sensor Network[D].Hehai University.2007
[7] YANG M,QIN Q Q. The Routing Protocol Based on
WSN[J].Computer
Engineering
And
Applications .2004.32:130-13
[8] Murata T ,Ishibuchi H. Performance evaluation of genetic algorithms for flowshop scheduling problems[C]. In: Proc of the 1st IEEE Conf. on Evolutionary Computation.
Orlando:IEEE Press,1994.812-817
4. Conclusion
Wireless sensor network technology has been used widely day by day; however the limited energy resource is one of the bottlenecks to limit its application. To enhance the robustness and accuracy of the information that the entire network can obtain, the monitoring areas of the sensor nodes should be overlapped to make certain redundancy of the data that the sensor nodes gathered. Therefore the data fusion processing is needed to reduce the redundant information. So far in the research field of wireless
[9] YU B, SUN B, WEN N etc.NS2 And Network
Simulation [M].Beijing:Posts & Telecom Press,2007.4.
70
Authorized licensed use limited to: Mississippi State University. Downloaded on November 27, 2009 at 15:22 from IEEE Xplore. Restrictions apply.
