Towards IoT; Comparison of RSS based indoor localization using supervised learning and trilateration in WSN

The emergence of Internet of Things (IoT) the networking of objects and sensors remotely plays an important role in many applications of smart home, elder care, transportation, medical, agriculture, energy management and etc. The wireless sensor network (WSN) as a subset of IoT, facilitates IoT based networking by their placement of sensors for monitoring and recoding. The Received Signal Strength Indicator (RSSI), as the quantitative indicator of Received Signal Strength (RSS), is mostly used on indoor localization by placing sensors in WSN. In this research we use to place Wi-Fi nodes in WSN for acquisition of RSSI remotely over on cloud architecture for IoT. Predominantly this research compares two position estimation approaches used with WSN for localization, which essentially a part of object tracking in indoor environment. The Artificial neural network (ANN) based Feed Forward Neutral Network (FFNN) and deterministic trilateration techniques are used for position estimation, and their results are compared to put forward the appropriate localization technique for indoor environment. The paper presents distinctly the FFNN is the most suitable two dimensional (2D) positioning model for indoor localization.