Desktop version

Home arrow Computer Science

  • Increase font
  • Decrease font


<<   CONTENTS   >>

EXPERIMENTAL VALIDATION

In this section, a detailed experimental validation takes place to verify the superior nature of the represented SS-KH algorithm under three unique scenarios. A set of evaluation parameters applied to measure the performance are energy efficiency and network lifetime.

Implementation Setup

The simulation takes place using MATLAB and a detailed experimental analysis is carried out with respect to three scenarios based on the distance to BS. To ensure the reliable

TABLE 8.1 Simulation Parameters

Parameters

Value

Area

100 x 100 m2

Location (Si)

50, 50

Location (S2)

100,100

Location (S3)

150, 50

E0

0.5 J

Node count

200

^elec

50 nj/bit

%

10 pj/bit/m2

^mp

0.0013 pj/bit/m4

Packet size

4000 bits

performance of the SS-KH algorithm, results are measured based on diverse locations of BS which are mentioned as follows:

a. Scenario 1 (SI): BS at middle of the target area

b. Scenario 2 (S2): BS at corner of the target area

c. Scenario 3 (S3): BS located far away from the target area

A sample WSN consists of 200 nodes which undergo random deployment in the sensing field of 100 x 100 m2. In addition, first-order radio energy model is considered. For comparative analysis, a set of five techniques, namely SS [28], LEACH [29], DEEC [30], TEEN [31], and FUCHAR [7], were taken. The assumption of the simulation parameters presented in the study is provided in Table 8.1. Besides, the deployment of nodes in all the three scenarios and the corresponding cluster construction is demonstrated in Figures 8.4 and 8.5 correspondingly.

Performance Analysis

Comparative Study on Energy Efficiency Analysis

To ensure the energy efficiency analysis of SS-KH algorithm, an investigation of average remaining energy takes place under three scenarios and the outcome is shown in Figures 8.6-8.8. The usage of energy is computed by the average energy consumed by every individual node. As seen in figure, it is evident that the introduced SS-KH algorithm shows more efficiency on comparing with other methods. It is due to the fact that efficient CH and proper cluster sizes are computed by the nature of SS-KH algorithm. The proper selection of CHs as well as size of the clusters minimizes the energy requirement in the entire WSN. In addition, the LEACH shows ineffective performance due to the nature of following characteristics: random CH selection and dedicated data transmission from CHs to BS. Simultaneously, TEEN offered minimum energy consumption over LEACH. However, it fails to achieve efficient results over the SS-KH algorithm. As TEEN is a threshold-based model, the decreased number of data transmission results in minimum energy utilization.

Deployment of nodes under three scenarios

FIGURE 8.4 Deployment of nodes under three scenarios.

However, the way of selecting CHs randomly results in maximum energy expenditure over the compared techniques. The DEEC exhibits somewhat manageable performance over the other methods.

However, the selection of TCH arbitrarily results in poor performance compared to SS-KH, SSUCP, and FUCHAR. Even though FUCHAR and SSUCP offer effective performance, they lead to inefficient outcome compared to the presented SS-KH algorithm. On all the applied different scenarios, the SS-KH algorithm showed extraordinary results over

Construction of clusters under three scenarios

FIGURE 8.5 Construction of clusters under three scenarios.

SI: Average energy consumption

FIGURE 8.6 SI: Average energy consumption.

the methods used for comparison purposes. Once the data communication has begun, the remaining energy level starts to reduce and it stuck to null at certain time. In that case, the node will be declared as dead node and the alive node count begins to deceases. The clustering technique whose node count is high after the execution of numerous rounds is termed as an efficient model.

S3: Average energy consumption

FIGURE 8.8 S3: Average energy consumption.

Comparative Study on Network Lifetime Analysis in Terms of Alive Nodes

From the various dimensions of validating the WSN lifetime, this study is based on the alive node count. From the view of deploying nodes, every node will stay alive. Figures 8.9-8.11 demonstrate the alive node count of diverse methods under three unique scenarios correspondingly. Similar to the average residual energy examination, minimum lifetime is

Network lifetime analysis of S2

FIGURE 8.10 Network lifetime analysis of S2.

obtained by the LEACH over the compared methods. At the same time, TEEN performs well compared to LEACH, however, it failed to show better results over the other methods. In the same way, the DEEC, SSUCP, and FUCHAR are found to be effective over the compared techniques; however, the SS-KH is an effective technique. The figures clearly show that the maximum lifetime is attained by the SS-KH algorithm.

Figure 8.12 and Table 8.2 show the investigation of the results attained by different methods in terms of FND, HND, and LND. From the table, under the SI, it is clear that the

Network lifetime analysis in terms of FND, HND, and LND

FIGURE 8.12 Network lifetime analysis in terms of FND, HND, and LND.

FND of the LEACH takes place by 53 rounds which its minimum network lifetime. Next, the TEEN provides better results than LEACH with the FND at 75 rounds. Simultaneously, the DEEC model somewhat lengthens the FND by 84 rounds. Next, the FUCHAR shows better lifetime with the FND of 110 rounds whereas the SS extends the FND to 170 rounds. However, the proposed SS-KH algorithm achieves maximum network lifetime with the FND of 198 rounds. Similarly, under S2 and S3, the proposed method achieves better performance in terms of FND. On measuring the results in terms of HND, under the SI, it is clear that the HND of the LEACH takes place by 996 rounds which its minimum network lifetime. Next, the TEEN provides better results than LEACH with the HND at 1001 rounds. Simultaneously, the DEEC model somewhat lengthens the HND by 1016 rounds. Next, the FUCHAR shows better lifetime with the HND of 1101 rounds whereas the SS extends the HND to 1046 rounds. However, the proposed SS-KH algorithm achieves maximum network lifetime with the HND of 1601 rounds. Similarly, under S2 and S3, the proposed method achieves better performance in terms of HND.

On measuring the results in terms of LND, under the SI, it is clear that the LND of the LEACH takes place by 1401 rounds which its minimum network lifetime. Next, the TEEN provides better results than LEACH with the LND at 1523 rounds. Simultaneously, the

TABLE 8.2 Comparison of Network Lifetime in Terms of FND, HND and LND

Methods

FND

HND

LND

SI

S2

S3

SI

S2

S3

SI

S2

S3

SS+KH

198

174

160

1645

1601

1595

2539

2460

2502

ss

170

131

124

1215

1046

1196

1900

1696

1642

FUCHAR

110

92

85

1101

1011

1006

1715

1646

1604

DEEC

84

72

65

1016

1001

989

1602

1421

1552

TEEN

75

60

40

1001

961

946

1523

1395

1501

LEACH

53

45

32

996

940

910

1401

1310

1420

DEEC model somewhat lengthens the LND by 1602 rounds. Next, the FUCH AR shows better lifetime with the LND of 1715 rounds whereas the SS extends the LND to 1900 rounds. However, the proposed SS-KH algorithm achieves maximum network lifetime with the LND of 2539 rounds. Similarly, under S2 and S3, the proposed method achieves better performance in terms of LND. These values proved that the SS-KH algorithm achieves maximum network lifetime over other methods under all the scenarios.

 
<<   CONTENTS   >>

Related topics