Bat Ant Lion Optimization-Based Generative Adversarial Network For Structural Heath Monitoring In IoT

Abstract

Attributable to the rapid growth of information technology, the Internet of Things (IoT) having strong permeability characteristics, huge usage of action and better comprehensive benefits. However, it encourages the development of IoT technology in the detection of structural engineering. Structural health monitoring (SHM) is responsible for identifying techniques and for prototyping systems performing a state diagnosis of structures. Its aim is to prevent sudden civil infrastructure failure as a result of several invisible sources of damage. This paper devises a novel method, namely bat-antlion Optimization dependent generative adversarial network (BALO-based GAN) for monitoring the states of structural health. Here, IoT nodes sense the signals of each channels and sensed data are transmitted to base station (BS) using Monarch-Earthworm (Monarch-EWA)-enabled secure routing protocol that selects the optimal path for the data transmission. After performing the IoT routing, the state of the structural health is monitored at the BS. For SHM, the input signal acquired from the IoT routing phase is fed to the pre-processing step for improving the signal quality for further processing. Then, the feature extraction is performed using fractional-amplitude modulation spectrogram (fractional AMS) for extracting the best features for improving the classification accuracy. The extracted features are adapted by the GAN, which is trained by BALO. The proposed BALO is newly designed by integrating the Bat algorithm and antlion optimizer. The proposed BALO-based GAN showed improved performance with maximal accuracy of 0.912, maximal sensitivity of 0.911, maximal throughput of 0.972 and maximal specificity of 0.913, respectively.