Localization in wireless sensor network using machine learning optimal trained deep neural network by parametric analysis

Abstract

Sensor nodes (SNs), which are a component of wireless sensor devices, have several uses in monitoring and surveillance. To perform location-based routing, the sensor nodes must be aware of their location reference (LR). The field of localization is the subject of several studies. Future developments in mobility, 3-D localization, energy preservation, and security are yet largely neglected. Here, optimization aided DNN trained with the measured distance based features like “Angle of Arrival (AoA) and Received Signal Strength Indicator (RSSI)” gives the predicted results on location of unknown nodes. To enhance the accuracy on localization, a hybrid model termed as Lion Assisted Firefly Algorithm (LAFA) model has been introduced. In this conference, a parametric analysis is made on the proposed algorithm by varying the parameter in LAFA. This includes the performance analysis of the model under each variation.