Energy Efficient Multi-Objective Task Allocation in Software-Defined Wireless Sensor Network
Abhishek Narwaria, Arka Prokash Mazumdar, Shubham Sharma · 2023
Traditional Wireless Sensor Networks (WSN) integrate the application in their sensor nodes, limiting their ability to fulfill additional functions. The application-specific nature of WSN restricts their scope and use-case. This results in the redundant deployment of other WSNs to meet various application-specific needs. To address the inflexible nature of WSN, a software-defined paradigm can aid in achieving flexibility, simplicity, management, and the ability to compute tasks concurrently. The proposed approach utilizes the characteristics of Software-Defined Wireless Sensor Networks (SDWSN) to provide energy-efficient task allocation from a basic application decomposition to task scheduling on sensor nodes while tracking latency with minimum task failure. We proposed a load-based resource allocation to balance the network energy through machine-bottleneck task scheduling to ensure minimal task failure. The simulation results demonstrate that our pro-posed method outperforms the state-of-the-art in the total energy consumption, task completion, task completion with varying node load, and percentage change in energy consumption.