Energy consumption minimization on LoRaWAN sensor network by using an Artificial Neural Network based application
Résumé
In this paper we use an application approach to minimize the energy consumption and increase the lifetime of a LoRaWAN sensor network. In this sensor network the nodes transmission cycles are controlled by an Artificial Neural Network (ANN) based algorithm. The algorithm predicts the data of nodes and by this method the nodes can avoid data transmission and stay more in Idle mode. For implementing this algorithm and for controlling the network we use the middleware MQTT, which is compatible with The Things Network public LoRaWAN server. With this control, the nodes can, for the best case, save up to 58.91% transmissions and extend their lifetime.
Mots clés
- data transmission
- minimisation
- wide area networks
- wireless sensor networks
- Downlink
- Neurons
- Switching circuits
- Servers
- nodes transmission cycles
- power aware computing
- Energy consumption
- Prediction algorithms
- Protocols
- LoRaWAN
- Artificial Neural Network
- Energy minimization
- Sensor Network
- Middleware
- neural nets
- middleware MQTT
- LoRaWAN sensor network
- energy consumption minimization