Adaptive Deep Rider LSTM-Enabled Objective Functions for RPL Routing in IoT Applications

Autor: A., Chaudhari D., E., Umamaheswari, Chaudhari, Dipalee, Umamaheswari, E.
Zdroj: International Journal of Information Security and Privacy; November 2021, Vol. 16 Issue: 1 p1-17, 17p
Abstrakt: This paper presents a proposed Objective Function (OF) design using various routing metrics for improving the performance of IoT applications. The most important idea of the proposed design is the selection of the routing metrics with respect to the application requirements. The various metrics, such as Energy, Distance, Delay, Link quality, Trust (EDDLT) are used for improving the objective function design of the RPL in various IoT applications. Here, the Adaptive Deep rider LSTM is newly employed for the energy prediction where the Adaptive Deep Rider LSTM is devised by the combination of the adaptive theory with the Rider Adam Algorithm (RAA), and the Deep-Long Short Memory (Deep-LSTM). However, the evaluation of the proposed method is carried out energy dissipation, throughput, and delay by achieving a minimum energy dissipation of 0.549, maximum throughput of 1, and a minimum delay of 0.191, respectively.
Databáze: Supplemental Index