Distributed RSS-Based 2D Source Localization System in Extended Indoor Environment
Tunguturi Sridher ,
Achanta Dattatreya Sarma ,
Perumalla Naveen Kumar and
Kuruva Lakshmanna
The evolution of computing and network technologies which involve thousands of devices that are connected wirelessly to serve variety of applications in Internet-of-Things (IoT) draws significant interest in locating the indoor objects. In our paper, we focus on developing a hybrid source positioning technique with off-the-shelf hardware modules. A rectangular corridor with a multipath environment is considered in our work. For better localization accuracy, the corridor is classified into segments with threshold RSS values. Based on the measurement data segment-wise logarithmic regression models are developed, and the performance in terms of Correlation Coefficient (R2) and Root Mean Square Error (RMSE) is evaluated. For localization, basically trilateration is used. However, to overcome the adverse issues due to the indoor environment such as flip ambiguity, uncertainty in range measurements, circumscribing the circle's scenarios, two circle intersection, dynamic circle contraction, and expansion methods are used. Relevant Pseudocode algorithms are presented. The proposed hybrid method significantly improves the localization accuracy. The standard deviation of errors in x and y directions are about 16.75 cm, 66.24 cm in the first segment and 19.75 cm, 60.16 cm in the second segment. The analysis and results are useful in establishing state of the art IoT and future generation 5G networks.