A Novel Learning Automata and Genetic Algorithm-based Energy Efficient and Reliable Data Delivery Mechanism in Wireless Sensor Networks

Wireless sensor networks (WSNs) pose many challenges- in spite of their crucial role in data transmission and reception covering wide range of applications. However, in adverse environmental conditions, wireless link instability prevents sensed data from reaching the sink node as well as leading to packet loss. In order to improve reliability of data delivery, retransmission and redundancy-based strategies can be employed. Nevertheless, these strategies lead to higher energy consumption with longer packet delivery delays. Therefore, ensuring energy efficiency and consistent delivery of data is of primary concern with WSNs. The proposed research work is predominantly focused on the route construction by selecting best resourceful nodes in terms of all the vital parameters of sensor nodes like residual energy, buffer space, distance & link quality (that leads to packet dropping). This work presents a routing mechanism for WSNs that leverages learning automata (LA) and genetic algorithm (GA). At first LA is hired to generate initial population. Further, GA is employed to establish energy efficient and reliable routes. Moreover, a novel fitness function is developed with due consideration of parameters like residual energy, buffer space, distance and link quality. Thus, this approach guarantees best possible parent chromosomes for the genetic operations. Simulation results depict that this algorithm condenses energy consumption, diminishes data delivery delays, decreases the number of data transmissions, while outspreading lifetime of network.