Abstract
Energy conservation is a crucial issue to extend the lifetime of wireless sensor networks (WSNs) where the battery capacity and energy sources are very restricted. Intelligent energy-saving techniques can help designers overcome this issue by reducing the number of selected sensors that report environmental measurements by eliminating all replicated and unrelated features. This paper suggests a Hybrid Sensor Selection (HSS) technique that combines filter-wrapper method to acquire a rich-informational subset of sensors in a reasonable time. HSS aims to increase the lifetime of WSNs by using the optimal number of sensors. At the same time, HSS maintains the desired level of accuracy and manages sensor failures with the most suitable number of sensors without compromising the accuracy. The evaluation of the HSS technique has adopted four experiments by using four different datasets. These experiments show that HSS can extend the WSNs lifetime and increase the accuracy using a sufficient number of sensors without affecting the WSN functionality. Furthermore, to ensure HSS credibility and reliability, the proposed HSS technique has been compared to other corresponding methodologies and shows its superiority in energy conservation at premium accuracy measures.
Highlights
Wireless Sensor Networks (WSNs) could be described as a group of connected sensor nodes [1], which are small-sized devices with restricted resources such as power supply and memory
Energy-efficient management is mostly conceptualized as mechanisms and strategies by which the overall energy of the network is allocated, arranged, and used effectively by all sensors so that the network remains fully functional for its expected life
This paper proposes an intelligent Hybrid Sensor Selection (HSS) technique for efficient-energy conservation in WSNs
Summary
Wireless Sensor Networks (WSNs) could be described as a group of connected sensor nodes [1], which are small-sized devices with restricted resources such as power supply and memory. In a traditional sensor network, every node should monitor physical environmental prerequisites like sound, temperature, pressure, humidity, motion, light, vibration, etc. The generated information from the sensor network is quite correlated, and reporting every individual sensor reading is a waste of energy resources. CMC, 2022, vol., no.3 of machine learning methodologies’ adoption in WSNs applications to learn about and discover correlated data, predictions, decisions, and information or data classification. The first reason is that sensor nodes aren’t operating as predicted due to sudden environmental behavior. The second reason is because of the unpredictable environments where WSNs are distributed
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
