The AI-DRM protocol to enhance the lifetime of wireless sensor network

Abstract

Energy is a major research challenge in wireless sensor networks since it is placed in an area that is inaccessible to humans. In the current study, nodes send data to their neighboring nodes at any distance using the same energy level. Smaller distances require less energy to transmit to adjacent nodes, creating a strong research gap. High-distance transmissions require more energy. The node must tailor its transmission energy to distance, not fixed energy. The best transmission power for communication is determined via the neural network-based machine learning technique, which is based on the propagation model and network properties, such as the node density, residual energy, and energy harvesting rate. In this work, sensor nodes transmit information to their neighboring nodes via the multiple linear regression model for dynamic radio tuning with the FRIIS propagation model, and the simulation records the node's energy consumption. Compared with the four recent best current methods that increase the W.S.N. lifetime, the proposed protocol is better and uses less power. The proposed AI-DRM protocol has sufficient residual energy to transmit the packet until 1403 rounds, which is higher than those of two recent energy-efficient protocols, the ARORA and the EACHS-B2SPNN protocols.

• 1.
  The AI-based dynamic transmission power protocol tunes the sensor nodes using a propagation model.
• 2.
  Prediction of lifetime of WSN
• 3.
  Effective utilization of all sensor nodes¹