Performance Analysis of Localization Techniques with Reduction in Location Error of Wireless Sensor Networks

Performance Analysis of Localization Techniques with Reduction in

Location Error of Wireless Sensor Networks

Bindhu Madhavi. P

1

Dr. M. Nagendra

2

1

Ph.D. Scholar

2

Professor

1,2

Department of Computer Science & Engineering

1,2

S.K.University, Anantapur, A.P

Abstract— The Basic issue of wireless sensor networks is Localization. Localization gives Positional information of sensor nodes. For obtaining location information of sensors we require Localization techniques (algorithms). For the deployment of sensor nodes, coverage and rescue operations location information is essential. Many applications such as routing and target tracking are all location dependent. Two Localization algorithms are used to obtain the node positions. Range Based & Range Free Localization techniques which are used to locate the nodes. Individually these two techniques are having limitations. To solve these limitations CDL (Combined and Differentiated Localization approach) method is proposed which uses DV -hop localization algorithm to calculate distance between the nodes. High Location accuracy is achievable by CDL method in reducing the location error. This paper is proposed with advances in Localization methods along with their location error results. The aim of this work is to find out the precise location of sensor node.

Key words: Wireless Sensor Networks (WSNS),

Localization, Localization Techniques, Average Location Error

I. INTRODUCTION

Sensor nodes are having restricted energies and possessions in wireless sensor networks. The most important aspect from wireless sensor network is the location information of sensor node. Localization techniques (algorithms) are used for obtaining location information of sensors. The geographical positions (location) of sensor nodes are find out through Localization techniques [1]. Initially Localization techniques estimate the locations of sensors by using the absolute position information [2]. The Global Positioning System (GPS) consists many localization approaches but they fails in indoors and under the ground and in forest. Two self Localization algorithms [3] are used to obtain the sensor node locations in WSN. Range Based & Range Free Localization techniques [4] which are used to compute the node locations. Range-based algorithm is used to compute the exact distance between the nodes and then use the information to localize the nodes [5]. Range based algorithms attain good localization results, with diverse kinds of techniques such as: Time of arrival (TOA) [7], Time difference on arrival (TDOA) [8], Angle of arrival [9], and Received signal strength indicator (RSSI) [6]. Range-free algorithm takes only network connectivity information between nodes or information of hop count to localize the nodes. The various Range-free techniques are centroid algorithm[10], DV-Hop algorithm [11,12] APIT algorithm [13]. The localization results offered by range-based approaches are more accurate than the range-free approaches but it requires additional Hardware which is expensive for implementation. Without any Hardware

requirements Range free localization algorithm achieves high accuracy which is an alternative cost effective approach [14]. In practical situation both the algorithm does not able to address the issue of the localization because of dynamic environments. To solve the limitations in Range based and Range free algorithm we propose the solution CDL, Combined and Differentiated Localization approach [16]. T h e p r o p e r t i e s o f range based and range free algorithms are inherited by CDL to overcome the limitations and to increase the Location accuracy. I n C D L d i s t a nc e i s calculated first by using the DV -hop localization algorithm, and secondly improved ranging quality and location accuracy is calculated by double filtration and calibration process which in turn reduces the Location error. The organization of this paper is as follows: Localization Trouble is presented in section II, Analysis of existed Localization Techniques is presented in section III. The proposed and improved method is introduced in section IV. Conclusion of the paper is given in the V section. Simulation and results are provided in Section VI.

[image:1.595.308.539.386.589.2]

II. LOCALIZATION TROUBLE

Fig. 1: Localization of WSN

Localization difficulty is caused due to the distinction between the expected co-ordinate and the real co-ordinate. The formula for the fault location is given by

2 ) ' ( ) '

( _{n n n n}

n x x y y

Error    

(1)

The real co-ordinates of node N are (xn, yn)

respectively and the expected co-ordinates of node N are

) ,

(x'n yn' respectively. The most important significant factor

from the localization is Average location error. The average error expressed in terms of percentage is as follows:





 N

n n

Error NR

or AverageErr

1

% 100 * 1

(2)

III. ANALYSIS OF EXISTED LOCALIZATION TECHNIQUES

The two self Localization Techniques [3] are Range Based & Range Free Localization techniques [4] which are used to compute the node locations.

A. Range-Based Algorithm:

It computes the correct distance between the nodes and then use the information to localize the nodes [5]. Range based approach does not require details about the connectivity or availability of the nodes due to hardware limitations in WSN devices. Several Range based techniques are:

 Time of arrival (TOA) [7]

 Time difference on arrival (TDOA) [8]  Angle of arrival [9], and

 Received signal strength indicator (RSSI) [6].

[image:2.595.45.282.181.405.2]

1) Time of Arrival:

Fig. 2: Time of Arrival

Time of arrival (TOA) or Time of flight (TOF), quantify the distance between the two nodes by using the travel time from the transmitter to the receiver.

In order to apply TOA, minimum three sensors are needed. If the distance between three sensors is known, then at the junction location can be found with three circles around every sensor with the radius which denotes distance measurement. Deficient calculations create area of uncertainties between each of these sensors in which the transmitter may contain

2) Time Difference of Arrival:

Multilateration, or Hyperbolic positioning, is the method used by Time difference of arrival (TDOA), which uses for node localization. TOA uses travel time from transmitter to the receiver to measure distances. In TDOA the transmitter does not required to be in sync with the sensor. So does not require the distance between the transmitter and the receiver. Synchronization is the only requirement between the sensors since the Localization calculation is based on diversities of time/distance. So in TDOA to find the distance between each sensor [19] different travel time diversities are used

Fig. 3: Time Distance of Arrival

3) Angle-of-arrival:

[image:2.595.370.482.193.306.2]

The angle between some reference direction and transmission direction of a random wave is recognized as orientation and is called as Angle of arrival. The orientation is a fixed direction where AOAs are estimated. From the north [20] in a clockwise direction this orientation is represented in terms of degrees. It uses the hop count between the nodes for measuring the distance; it uses geometric relations to estimate the received signals angle and to estimate the locations of nodes. AOA is linked to direction of arrival (DOA) which can be expected by the virtual or complete angles between the neighbors.

Fig. 4: Angle-of-arrival

4) Received Signal Strength Indicator:

RADAR [17] or RSSI technology is proposed for hardware controlled systems. These techniques are used to translate signal strength into distance estimates. The distances between two nodes are estimated by using RSSI based on the strength in the received signal through another node. The strength of the signal becomes less due to large distance towards receiver node. Tentatively the strength of the signal is inversely proportional to squared distance; along with a known radio propagation model may be used to convert the signal strength into distance.

(3)

Fig. 5: Received Signal Strength Indicator Method [17] The problems such as uneven signal propagation noises, obstacles, type of antenna, and multi-path fading, from RF systems [17][18], makes range measurements imprecise. A two-stage improvement positioning and calibration constraint is planned to reduce error to with some tolerable limit. So we generate a system calibration, in which the values of RSSI and distances are evaluated at the forefront of time within a restricted environment. The main advantage is cheap in cost, due to capable receivers.

B. Range-Free Algorithm:

[image:2.595.308.555.425.594.2] [image:2.595.102.231.647.762.2]

Localization. So, these techniques are less expensive and require less information compared to range-based algorithms.

The requirements for these algorithms are  Which nodes are within radio range?  Node estimated locations.

 Sensors radio range The Range-free techniques are

 Centroid algorithm[10],

 Approximate point in triangle algorithm[13],  DV-Hop algorithm [11,12]

C. Centroid Algorithm:

To estimate the location of sensor node [10] this technique uses the connected anchor node position (Xi, Yi). A signal with location information is transmitted by the anchor nodes. The positions of sensor nodes (Xest, Yest) are computed as centroid of the positions of all the neighboring connected anchor nodes to it as:

The estimated position of the sensor node is represented as Where, (Xest, Yest) and N is the number of neighboring connected anchor nodes to the sensor node. This method is simple and financial, but the estimated location results of this method are very poor and localization error is very high, which is improper for many applications.

Fig. 6: Centroid Localization Algorithms

D. Approximate Point in Triangle Algorithm (APIT):

[image:3.595.344.508.549.748.2]

It is a area based range free localization method in which anchor nodes known their positions based on the self-equipped high power transmitters. APIT is located in an area In order to perform position estimation the approximate point should locate in an area by dividing that area into triangular zones between anchor nodes

Fig. 7: Area Based APIT Algorithm

Each node‟s presence inside or outside the triangular regions allows decreasing the feasible location until all the possible sets have reached to an acceptable accuracy. The point in triangulation test (PIT) is used to narrow down the possible area that a target node resides. According to this test, if the node is not inside a triangle, it

requires moving [13]. In order to estimate the distance it uses Signal Strength. It just assumes that signal strength decreases steadily with the distance.

E. DV Hop Algorithm:

Distance Vector Hop algorithm[11,12] is proposed by Dargos Niculesu and Badri Nath. It is a disseminated hop by hop positioning algorithm.The execution of the program involves three steps.After hop count values to all anchor nodes.In the next step,when the anchor assigns hop count value to other anchors,average size of one-hop is approximated.Then the approximated average is conveyed to the whole network.The average size is calculated using the formula:

Hopsizei =





     i j ij i

j i j i j

h

y y x

x ₎2 _{( )}2

(

(5)

The coordinates of i and j are (xi ,yj) and

(xj,yi)respectively. hij are the hops between anchor i and

anchor j. After all mysterious nodes have acknowledged the hop-size from anchor nodes which have the minimum hops between them; they calculate the distance to the anchor nodes based on two factors: size of the hop and least hop

count

h

_id, using the following formula:



i

d

h

_id* HopSize_i(6)

From second step, the position of the mysterious nodes is measured according to the distance with each anchor node. The coordinates of the mysterious node is (x, y), and the coordinates of anchor i is (

x

_i,

y

_i). Assume

d

_iis

the distance between anchor nodes to mysterious nodes, then we have the following formula:

                     2 2 2 2 2 2 2 2 2 2 1 2 1 2 1 ) ( ) ( . . ) ( ) ( ) ( ) ( n n

n y y d

x x d y y x x d y y x x (7)

The above formula can be introduced with the subsequent linear equation:

AP = B. Where

P = _      y x (8)











































 n n n

n n n n n

y

y

x

x

y

y

x

x

y

y

x

x

A

1 1 2 2 1 1

..

...

..

...

*

2

(9)                                     

 21 2

2 2 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2 1 . . . n n n n n n n n n n n y y x x d d y y x x d d y y x x d d B (10)

B A A A

P T 1 T

)

( 

 ₍₁₁₎

[image:4.595.106.229.118.228.2]

The least number of hops and the distance that separates itself from each anchor can be obtained by the standard nodes by using the first two steps of DV-hop algorithm.

Fig. 8: DV Hop Localization Algorithms

IV. PROPOSED WORK

The limitations from Range based and Range free methods are overcome by using CDL, Combined and Differentiated Localization approach [16]. It consists of virtual-hop localization, local filtration, ranging quality aware calibration. Initially it calculates distance between the nodes by using the DV -hop localization algorithm, and then it improves the range quality and location accuracy process by double filtration and calibration process. DV-hop algorithm results a big error due to the, random distribution of nodes. To overcome this problem we propose an algorithm [21] based on the expected distance to the nearby anchor. An [22], improved DV-Hop algorithm consists of selective Three-anchor DV-hop algorithm, based on the connectivity vector. In this paper the development of algorithm is specified through two steps in which a mysterious node can calculate its position by selecting three anchors with reference to the nearest one.

Fig. 9: CDL workflow

Virtual Hop Localization use to calculate the distance using Range free algorithm, Local Filtration picks good nodes with good neighbors. Calibration finds the node with the best range measurement.

A. Virtual-Hop Localization:

Virtual-hop localization [16] calculates the node locations first. It, counts the virtual hops Each node has to count virtual hops to resolve the errors. Every sensor can be referred as a node in a graph and all the nearby nodes are connected, communicated with each other in one hop through their edges only.

B. Model-based Filtration:

Filtration [23] is used to find good nodes with high

localization accuracy. A good node is likely to have both good and bad links. It consists of two filtering methods used to identify good nodes. The good nodes are whose location accuracy is satisfactory. The two local filtration techniques are neighborhood hop-count matching and neighborhood sequence matching.

C. Neighborhood Hop-Count Matching:

Neighborhood hop-count matching [24] filters the bad nodes. Here each node takes neighborhood hop-count matching as the first step to identify whether it is a good node based on local connectivity information. Note that hop count is indeed a rough estimation of the distance between two nodes. If a node‟s hop counts to its neighbors greatly mismatches the distances calculated using the nodes‟ estimated coordinates, w.h.p. the local node‟s coordinates will have a large error. First, every node exchanges the estimated coordinates with its 2-hop neighborhood. Second, after received the estimated coordinates, estimates the distance between them. Third, for each node within its 2-hop neighborhood, estimates the hop count to its communication range.

D. Neighborhood Sequence Matching:

Neighborhood sequence matching[24] distinguishes good nodes from bad ones based on RSSI sequence and distance sequence. Though model-based straightforward filtration is infeasible, RSSI still offers useful information. Generally, the RSSI between two nodes decreases monotonically as the distance increases. First, sorts its neighbors in descending order with regard to the RSSI from them, generating a sequence number for each neighbor. By mapping the sequence numbers we get the first sequence called RSSI sequence. Second, according to the estimated coordinates, sorts its neighbors in the ascending order with regard to the estimated distance between them, generating the second sequence is called distance sequence.

E. Ranging-Quality Aware Calibration (RQAC):

Calibration [23] [24] plays an important role in finding a node with better ranging quality. Given the range a measurement between bad node and its good neighbors, the estimation off‟s location usually works by minimizing an objective function. RQAC Estimator is used to estimate the ranging quality of a good node with its good neighbors. As the set of undetermined nodes includes both good and bad, we only use good nodes as references and do not include any undetermined nodes in the calibration. a threshold. For RQAC estimator, the growth trend is restrained by decreasing ranging quality

V. COMPARISONS OF ERRORS:

In the normal dv-hop the range of error is 8.7 and the snapped inducing shaped residuals is 5.6 and the combined and differential localization with triple anchor based DV hop has the error 2.2 and ongoing process carbon sink gives still less range of errors.

VI. CONCLUSION

[image:4.595.71.264.471.595.2]

and the emission measurements and there will be forest fire risk prediction, by using this approaches we can get the higher accuracy, efficiency and the consistent performance in the remote areas and from this technique there many of them may government organization and non-government organization and from not only we can find the changes in the environmental factors undergoing in wild areas and also climatic changes happening in the areas like wild and also it can use in the remote areas to check the unknown person or any wild animals entering into the remote areas can easily find and we can protect people from the various kinds of human and wild attacks through deploying the wireless sensor network

REFERENCES

[1] H. Dai, A. G. Chen, X. F. Gu, and L. He, “localization algorithm for large-scale and low-density wireless sensor networks,” Electronics Letters, vol. 47, no. 15, pp. 881–883, 2011.

[2] G. Mao, B. Fidan, and B. D. O. Anderson, “Wireless sensor network localization techniques,” Computer Networks, vol. 51, no. 10, pp. 2529–2553, 2007. [3] F. B. Wang, L. Shi, and F. Y. Ren, “Self-localization

systems and algorithms for wireless sensor networks,” Journal of Software, vol. 16, no. 5, pp. 857–868, 2005. [4] G. Q. Gao and L. Lei, “An improved node localization

algorithm based on DV-HOP in WSN,” in Proceedings of the IEEE International Conference on Advanced Computer Control (ICACC ‟10), vol. 4, pp. 321–324, March 2010.

[5] H. Chen, K. Sezaki, P. Deng, and C. S. Hing, “An improved DV-hop localization algorithm for wireless sensor networks,” in Proceedings of the 3rd IEEE Conference on Industrial Electronics and Applications (ICIEA ‟08), pp. 1557–1561, Singapore, June 2008. [6] T. S. Rappapport, Wireless Communications:

Principles andPractice, Prentice Hall, Upper Saddle River, NJ, USA, 1996.

[7] Girod and D. Estrin, “Robust range estimation using acoustic and multimodal sensing,” in Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 1312–1320, November 2001. [8] X. Cheng, A. Thaeler, G. Xue, and D. Chen, “TPS: a

time-based positioning scheme for outdoor wireless sensor networks,” in Proceedings of the 23rd IEEE Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM ‟04), pp.2685– 2696, Hong Kong, China, March 2004.

[9] D. J. Torrieri, “Statistical theory of passive location systems,” IEEE Transactions on Aerospace and Electronic Systems, vol. 20, no. 2, pp. 183–198, 1984. [10]N. Bulusu, J. Heidemann, and D. Estrin, “GPS-less

low-cost out- door localization for very small devices,” IEEE Personal Communications, vol. 7, no. 5, pp. 28– 34, 2000.

[11]D. Niculescu and B. Nath, “Ad hoc positioning system (APS),” in Proceedings of the IEEE Global Telecommunicatins Conference (GLOBECOM ‟01), pp. 2926–2931, San Antonio, Tex, USA, November 2001. pp. 267–280, 20034,.

[12]D. Niculescu and B. Nath, “DV based positioning in Ad hoc net- works,” Telecommunication Systems, vol. 22, no. 1––4, pp. 267–280, 2003.

[13]T. He, C. D. Huang, B. M. Blum, J. A. Stankovic, and T. Abdelza- her, “Range-free localization schemes for large scale sensor netowrks,” in Proceedings of the 9th the Annual International Conference on Mobile Computing and Networking, pp. 81–95, ACM, San Diego, Calif, USA, 2003.

[14] T. He, C. Huang, B. M. Blum, J. A. Stankovic, and T. Abdelzaher, “Range-free localization schemes for large scale sensor net- works,” in Proceedings of the 9th Annual International Confer- ence on Mobile Computing and Networking (MobiCom ‟03), pp.81–95, San Diego, Calif, USA, 2003.

[15]Y. Shang, W. Ruml, Y. Zhang, and M. P. J. Fromherz, “Local- ization from mere connectivity,” in Proceedings of the 4th ACM International Symposium on Mobile Ad Hoc Networking & Com- puting (MobiHoc ‟03), pp. 201–212, Annapolis, Md, USA, 2003.

[16]Jizhong Zhao and Yuan He,” Localization of the wireless sensor networks in wild:pursuit of ranging quality”, in Proc. ACM SenSys, 2013, pp.281–294. [17]J. Hightower, G. Boriello and R. Want, SpotON: An

indoor 3D Location Sensing Technology Based on RF Signal Strength, University of Washington CSE Report #2000-02-02, February 2000.

[18]P. Bahl and V. N. Padmanabhan, RADAR: An In-Building RF-Based User Location and Tracking System, In Proceedings of the IEEE INFOCOM „00, March 2000.

[19]R. Exel and P. Loschmidt, ―High Accurate Timestamping by Phase and Frequency Estimation‖, in ISPCS 2009 International IEEE Symposium on Precision Clock Synchronization for Measurement, Control and Communication, Brescia, Italy, 2009, pp.126-131.

[20]L. Gui, A. Wei, and T. Val, “A two-level range-free localization algorithm for wireless sensor networks,” in IEEE Conference on Wireless Communications Networking and Mobile Computing, pp. 1–4, 2010. [21]D. Pointcheval and J. Stern, “Security arguments for

digital signatures and blind signatures,” Journal of Cryptology, vol. 13, no. 3, pp. 361–396, 2000.

[22]L. Harn and Y. Xu, “Design of generalized ElGamal type digital signature schemes based on discret logarithm,” Electronics Letters, vol. 30, no. 24, pp. 2025–2026, 1994

[23]Y. Shang, W. Rumi, Y. Zhang, and M. Fromherz, “Localization from connectivity in sensor networks,” IEEE Trans. Parallel Distrib. Syst.,