Fault detection in single-hop and multi-hop wireless sensor networks using a deep learning algorithm

Abstract

The wireless sensor network(WSN)has received significant recognition for its  positive  impact  on  environmental  monitoring,  yet  its  reliability  remains prone  to  faults. Common  factors  contributing  to  faults  include  connectivity loss  from  malfunctioning  node  interfaces,  disruptions  caused by  obstacles, and increased packet loss due to noise or congestion. This research employs a  variety  of machine  learning  and  deep  learning techniques  to  identify  and address these faults, aiming to enhance the overall lifespan and scalability of the WSN. Classification  models  such  as support  vector  machine  (SVM), gradient  boostingclasifer  (GBC),  K-nearest  neighbours(KNN), random forest, and decision tree were employed in model training, with the decision tree emerging as the most accurate at 90.23%. Additionally, a deep learning approach, the  recurrent  neural  network(RNN),  effectively  identified  faults in sensor nodes, achieving an accuracy of 93.19%.