Performance evaluation of different data aggregation algorithms for different types of sensors in WSN based cluster

In wireless sensor networks, data aggregation algorithms are used to extend the network lifetime Size of data transmission from the cluster head node to base station show a critical role in CH nodes energy consumption. In this project three different data aggregation algorithms coding schemes based...

Full description

Saved in:
Bibliographic Details
Main Author: Ali, Wala'a Hussein
Format: Thesis
Language:English
English
English
Published: 2018
Subjects:
Online Access:http://eprints.uthm.edu.my/456/1/24p%20WALA_A%20HUSSEIN%20ALI%20AL-KAMIL.pdf
http://eprints.uthm.edu.my/456/2/WALA_A%20HUSSEIN%20ALI%20AL-KAMIL%20COPYRIGHT%20DECLARATION.pdf
http://eprints.uthm.edu.my/456/3/WALA_A%20HUSSEIN%20ALI%20AL-KAMIL%20WATERMARK.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In wireless sensor networks, data aggregation algorithms are used to extend the network lifetime Size of data transmission from the cluster head node to base station show a critical role in CH nodes energy consumption. In this project three different data aggregation algorithms coding schemes based relative difference (CS-RD), an adaptive method of data aggregation that exploits the spatial correlation between the sensor nodes (ADAM) and coding schemes based the factor of precision (CS-FP) are evaluated. The compared three algorithms are implemented on 15 different scenarios. The algorithms applied separately with (1) Mean (2) Median (3) Mode (4) Geometric mean (5) Harmonic mean. Each scenario applied separately for various sensors temperature, humidity, light and voltage. The performance metrics studied are energy consumption, average of absolute error and data compression ratio. The simulation results show that the best performance is shown by the CS-RD algorithm. The ADAM show intermediate performance for all sensors. In the average result, it can be said that the accuracy of CD-PF is better than other algorithms. But it shows the worst performance in term of energy consumption and data compression ratio for all scenarios. From the results its observed that select the mechanism to determine the central point effect in three aggregation algorithms performance. For temperature and humidity sensors the best performance is in term energy consumption by applied CSRD with all methods which was below 800 uJ and the compression ratio above 90% with acceptable error. In average CS-PF show the worst performance especially with Mode method. For voltage sensor in this study, The ADAM and CS-FP with Mean /Gmean /Hmean methods showed the better performance in energy consumption below 1000 uJ, the compression ratio about 91% and acceptable error. Also CS-RD show acceptable performance which is above 85% for compression ratio which that reflect the energy consumption. The Mode method effect negatively in performance all algorithms, the worst CS-RD with Mode in term of energy consumption closes to 2500uJ. Finally, for light sensor the best performance showed by applied CS-RD algorithm with all central point methods, where the energy consumptions below 1400 uJ. The CS-PF/ADAM with Mode showed the highest energy consumptions above 4200 uJ for CS-PF and above 2400 uJ for ADAM.