DEVELOPMENT OF INDOOR LOCALIZATION SYSTEMS USING VISIBLE LIGHT COMMUNICATION

Abstract

Visible light communication (VLC) has recently been the focus of research due to its widespread applications. This thesis practically demonstrates how VLC can be used for: (i) Indoor localization utilizing light emitting diodes (LEDs) and (ii) Secure and reliable communication. In the domain of indoor localization, the author has developed four algorithms namely (i). Indoor positioning using k-means clustering (ii). Indoor positioning using beam scanning (iii). Indoor positioning based on received signal strength and bi-literation and (iv). Achieving localization using a heuristic approach. All the developed techniques are based on LEDs. The first two approaches have been validated through experiments and the results show an accuracy of tens of centimeters (37 cm for the first technique and up to 13 cm for second technique) is achieved which are comparable to the state of the art indoor localization techniques available in the literature. Furthermore, to the best of author’s knowledge, they are the first one to introduce machine learning to achieve localization for VLC. Security is brought into the developed VLC system through spatial diversity based transmission using two optical transmitters and the reliability in the link is achieved by a newly proposed method for the construction of structured parity check matrix for binary Low Density Parity Check (LDPC) codes. Experimental results show that a successful secure and reliable link between the transmitter and the receiver can be achieved by using the proposed novel technique.