Energy Efficient-based Sensor Data Prediction using Deep Concatenate MLP
Made Adi Paramartha Putra, Ade Pitra Hermawan, Dong‐Seong Kim, Jae‐Min Lee
Abstract
This paper proposes a system to reduce sensor energy consumption by predicting the next sensor value. The current implementation of the smart factory utilizes wireless sensor network nodes to monitor the environmental condition in real-time. Instead of periodically exploiting those nodes, a deep learning prediction-based algorithm is proposed in the cluster head to reduce sensing times and increase sensor lifetime. The cluster head can learn the behavior of each sensor nodes based on its previous value. The proposed scenario can be combined with existing solutions in sensor failure detection and recovery to provide a robust solution in the industrial environment.
Topics & Concepts
Wireless sensor networkComputer scienceEnergy consumptionReal-time computingFactory (object-oriented programming)Energy (signal processing)Computer networkEngineeringElectrical engineeringProgramming languageMathematicsStatisticsEnergy Efficient Wireless Sensor NetworksAir Quality Monitoring and ForecastingDistributed Sensor Networks and Detection Algorithms