Gentle Adaboost Multivariate Logistic Regression Based Energy Efficient Data Delivery In Manet

  • D.Sathiya and Dr.S.Sheeja

Abstract

            MAENT comprises mobile nodes (MN) distributed arbitrarily with no infrastructure. In MANET, transfer of data to the destination becomes difficult due to its dynamic topology. In order to overcome this issue, Gentle AdaBoost Multivariate Logistic Regression Based Data Delivery (GAMLR-DD) method is introduced in MANET. The GAMLR-DD Method comprises two major processes namely energy efficient node classification and link quality measure for improving the performance of data delivery with minimal delay. Initially, Gentle AdaBoost ensemble classifier is used to classify the energy efficient MN for efficient data transmission. The ensemble classifier uses ‘n’ number of weak learners (i.e., base classifier) for performing the energy efficient MN classification. The weak learners’ outputs are combined offering the final output of boosted classifier. After the MN classification, multivariate logistic regression is used to analyze and estimate the neighboring mobile node to attain efficient data delivery with better link quality. The link quality of the node is estimated based on the received signal strength, distance and bandwidth availability. The neighboring node with better link quality is considered for data transmission. In this way, packet delivery ratio (PDR) and network lifetime get improved. Simulation is carried out with different metrics, namely EC, PDR, packet loss rate (PLR) and end to end delay (EED). The observed simulation outcomes demonstrate that GAMLR-DD method effectively enhances the data packet (DP) delivery and decreases the EC, PLR as well as EED better than the conventional methods.

Published
2020-05-01
How to Cite
D.Sathiya and Dr.S.Sheeja. (2020). Gentle Adaboost Multivariate Logistic Regression Based Energy Efficient Data Delivery In Manet. International Journal of Advanced Science and Technology, 29(06), 5004 - 5016. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/19433