evaluating the impact of environmental conditions on wireless sensor nodes using stochastic reward nets

Wireless sensor networks (wsns) are one of the most important distributed computing systems and environmental conditions have a great impact on their functionality. some sensor nodes in wsns have a battery as the power source and use renewable energy such as solar energy to charge it. if the batteries are not charged by harvesting energy from the environment, the tasks of the sensor nodes will fail. to prevent it, the sensor nodes can also decide to migrate tasks to neighbor nodes based on their battery status. on the other, the arrival rate of tasks at day hours is more than the arrival rate of tasks at night hours, but the charging rate of batteries is higher during the day than at night. therefore, decisions of wsns should be based on information from environmental conditions. the different arrival rates of tasks and charge rates of the batteries at day and night hours as the main environmental conditions have been ignored in the modeling of wsns. in this paper, we model a wsn node using stochastic reward nets (srn) and then compute the steady-state probabilities of processing, failure, and migration of tasks and evaluate the impact of different environmental conditions on them in the wsns. the results prove that changing the charge rate has a greater impact on the wsn functionality than changing the arrival rate.