Stochastic modeling and availability optimization of wireless sensor network through particle swarm optimization

The availability optimization of a Wireless Sensor Networks (WSNs), used in agriculture, through the combination of stochastic modeling via the Markov process and Particle Swarm Optimization (PSO) is presented. WSN hold paramount importance across various complex industries and have numerous real-world applications. The reliability of WSN is essential for their successful application in agriculture, especially to cope up with rapid changes in environmental conditions. A continuous-time Markov chain is employed to develop a mathematical model for WSN by considering its critical subsystems’ failure and repair rates, such as the microcontroller, power unit, and sensing unit. Availability matrix for the subsystem highlights that the power unit requires more maintenance attention than other components. PSO is utilized to optimize availability function by an optimum selection of population size and iterations. The optimization process improved system availability from 0.9606 to 0.9946 for a population size of 15 and to 0.9935 at 28 iterations, demonstrating an approximate 3% enhancement based on optimized failure and repair parameters. The findings indicate that PSO is a valuable method for enhancing the availability of WSNs, which is essential for agricultural applications where system dependability directly influences productivity. Result gives deeper insights about optimizing WSN’s performance and establishing maintenance priorities for efficient agricultural operations.