DR-NAP: Data reduction strategy using neural adaptation phenomenon in wireless sensor networks

Internet of Things (IoT) has got significant popularity among the researchers' community as they have been applied in numerous application domains. Most of the IoT applications are implemented with the help of wireless sensor networks (WSNs). These WSNs use different sensor nodes with a limited battery power supply. Hence, the energy of the sensor node is considered as one of the primary constraints of WSN. Besides, data communication in WSN dissipates more energy than processing the data. In most WSNs applications, the sensed data generated from the same location sensor nodes are identical or time-series/periodical data. This redundant data transmission leads to more energy consumption. To reduce the energy consumption, a data reduction strategy using neural adaptation phenomenon (DR-NAP) has been proposed to decrease the communication energy in routing data to the BS in WSN. The neural adaptation phenomenon has been utilized for designing a simple data reduction scheme to decrease the amount of data transmitted. In this way, the sensor node energy is saved and the lifetime of the network is enhanced. The proposed approach has been implanted in the existing gravitational search algorithm (GSA)-based clustered routing for WSN. The sensed data are transmitted to CH and BS using DR-NAP. Real sensor data from the Intel Berkeley Research lab have been used for conducting the experiments. The experiment results show 47.82% and 51.96% of improvement in network lifetime when compared with GSA-based clustered routing and clustering scheme using Canada Geese Migration Principle (CS-CGMP) for routing, respectively.