Abstract: The devices included in IoT networks have sensors and actuators for monitoring their
surroundings. These operate on battery energy, according to the characteristics of the environment
in which they are deployed. To enhance the longevity of IoT networks, the devices need to avoid
any unnecessary sensing operations in order to reduce the power consumption rate. However, as
existing sensing methods use a fixed sensing period policy, battery power wastage is inevitable.
In this paper, a smart sensing period policy is proposed for efficient energy consumption in an IoT
network. The proposed method uses a learning model based on a back-propagation neural network.
Within the target time, it can efficiently use the battery energy without any surplus or wastage in the
quantity of preserved battery energy. In experiments, our proposed method shows improved results
in battery energy consumption rates compared to the existing sensing period methods.

Keywords: internet of things; smart sensing period; neural network; power consumption rate;
battery energy