A Internet of Things Improvng Deep Neural Network Based Particle Swarm Optimization Computation Prediction Approach for Healthcare System

Abstract

Internet of Things (IoT) systems tend to generate with energy and good data to process and responding. In internet of things devices, the most important challenge when sending data to the cloud the level of energy consumption. This paper introduces an energy-efficient abstraction method data collection in medical with IoT-based for the exchange. Initially, the data required for IoT devices is collected from the person. First, Adaptive Optimized Sensor-Lamella Zive Welch (AOSLZW) is a pressure sensing prior to the data transmission technique used in the process. A cloud server is used data reducing  the amount of data sent from IoT devices to the AOSLZW strategy. Finally, a deep neural network (DNN) based on Particle Swarm Optimization (PSO) known as DNN-PSO algorithm is used for data sensed result model make decisions based as a predictive to make it. The results are studied under distinct scenarios of the presented of the performance for AOSLZW-DNN-PSO method, for that simation are studied under different sections. This current pattern of simalation results indicates that the AOSLZW-DNN-PSO method is effective under several aspects.