STUDY ON EMBEDDED SENSOR NETWORKS

Abstract

Wireless sensor networks (WSNs) are widely used in a range of applications in the real w orld. WSNs have demonstrated their applicability in the field of medical, military, and broad range of surveillance applications and monitoring. This thesis focuses on three major aspects related to WSN namely Energy efficiency, Security and real time application of WSN. The contributions are Energy efficiency in wireless sensor networks (WSN) is important as the sensor nodes are operated with restricted battery life. The choice of modulation scheme and error control code play a significant role in energy consumption of WSN. The analysis shows that the energy consumption in WSN may be reduced by selecting optimal combination of modulation scheme and error control codes. Our simulation results show that by using BPSK modulation with Reed Solomon (RS) code saves 48 % energy in WSN at an internode distance of 60 meters in comparison to other modulation schemes. In general, wireless sensor networks comprises of large number of sensor nodes. The communication among them depends on behaviour of each sensor node. Hence for securing the network, it is necessary to identify malicious nodes. In this thesis we present various techniques for malicious node detection and prediction. Home automation is one of the applications of WSN. We present low cost, compact and flexible ZigBee based home automation for remote control of house hold devices. This system assures optimal usage of electricity, thereby reducing carbon footprint. we propose an intelligent auto-dipping system that will be placed on the dashboard of a vehicle. Once headlight of an oncoming vehicle is detected, the vehicle's high beam will be automatically dipped, with a dipper flash to signal the approaching vehicle. The v system can be most successful when it is mounted in every vehicle. This system can help to avoid accidents caused by the dazzling created by oncoming high beams. The proposed system uses a USB web cam and Raspberry Pi for real time processing. The validation of the system has been done using both simulation and actual hardware.