A Federated Texture Learning and Scheduling (FTLS) for Energy Efficient Data Aggregation Model in WSN System

Abstract

In the Wireless Sensor Network (WSN) system, the sensor data aggregation and the dissemination are major key factor that needs to consider for the effective sensor data transmission over the sensor nodes. For that, the statistical parameters from the sensor data that are captures by the sensor deices in different applications. The wireless sensors are sending their parameters to the sink for further analytical process and for the sensor data aggregation. The main features of this paper is to analyse the different clustering models for the sensor data feature learning models to enhance the texture based learning model by using Federated Texture Learning and Scheduling (FTLS). This also improves the data preprocessing, aggregation and clustering, based on the feature learning and scheduling of cluster management. This leads to energy efficient clustering model and the data aggregation model in WSN network system. Typically, the sensor data prediction and the arrangement becomes the critical issue in the industrial communication systems based on the size of data arguments. Considering of this, the proposed work intends to develop an optimization model for reducing the dimensionality of sensor data with improved classification performance. Related to that the machine learning based clustering technique is to develop the data arrangement with better performance rate in terms of statistical analysis and reduced time complexity factors. The experimental result justifies the performance of proposed work by comparing the existing methods by using validation of parameters of statistical analysis such as Sensitivity, Precision, F-Score and Accuracy.