Open AccessIntegration of Machine Learning Techniques in Virtual Wireless Sensor Network for insect monitoring
Author(s) -
Mohammad Equebal Hussain,
Rashid Hussain
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1998/1/012031
Subject(s) - wireless sensor network , computer science , sleep mode , node (physics) , controller (irrigation) , battery (electricity) , mode (computer interface) , real time computing , transmitter , sensor node , wireless , power (physics) , embedded system , key distribution in wireless sensor networks , wireless network , computer network , engineering , telecommunications , human–computer interaction , biology , agronomy , channel (broadcasting) , physics , structural engineering , power consumption , quantum mechanics
Agriculture industries are comparatively slow in adopting emerging technologies than any other industries despite lot of exciting research. The use of Wireless sensor network (WSN) is very important role in agriculture for more productive and sustainable growth. The structure of WSN is tightly application dependent. Every WSN have sensors, processing unit, low frequency radio wave transmitter and power supply using battery. With increasing number of interconnected WSN devices, there is substantial increase in data generation. It contains both control messages as well as application dependent data, collected by the sensors. The collected data are frequently sent to the nearest centralized controller for further processing and decision making. For continuous functioning of WSN, uninterrupted power supply is needed. Many researches are carried out to overcome these challenges. In this manuscript we are proposing a simple and effective machine learning techniques combined with pause and play method to increase battery life of WSN. This could be achieved in three stages play-pause-play (PPP) model. First by gathering data for some time (play mode), Second by putting WSN to sleep (pause mode), in the backend, apply machine learning algorithm that helps to build model from training data to predict the future data. At the final stage, put WSN back to play mode. Compare the result with actual data and fine tune the model to reduce the error. Using this method WSN will get enough sleep time to increase overall life by simulating the normal behaviour of sensor node. The sleep time will be calculated dynamically.