Physical Layer Security and Channel Estimation in Optical Generalized Index-Modulated Ofdm System

Abstract

Visible light communication (VLC) is a promising and not completely discovered technology for next-generation wireless communication systems. In this thesis, gen eralized LED index modulation (GLIM-OFDM), which is a new modulation method for VLC systems based on multi-input multi-output (MIMO) orthogonal frequency division multiplexing (OFDM), is used. In the GLIM-OFDM system, the real and imaginary components of the complex time-domain OFDM signals are first sepa rated from each other, and then the resulting bipolar signals are transmitted over a MIMO-VLC channel by encoding the signal information in light-emitting diode (LED) indices. Higher spectral efficiency and power efficiency are achieved by en coding the beacon information of complex signals to the position of the transmitter LEDs. As in RF systems, physical layer security (PLS) in VLC systems is an issue that must be taken into account to ensure the security of the system and prevent access to the transmitted information by unauthorized users. While examining PLS techniques for VLC systems, this thesis proposes a new and unique PLS al gorithm and channel estimation method for GLIM-OFDM systems. We consider Bayesian(MMSE) estimation to handle the estimation problem, and then we pro pose a further enhancement of the PLS algorithm with secrecy key generation. From the curves we obtained, it is shown that the algorithm provides a high secrecy rate for the VLC system, i.e., when the legitimate user takes the transmitted information correctly, the eavesdropper cannot reach the correct information but a distorted one. Channel information also has an important place for proper transmission. In this study, the mean square error (MSE), bit error rate (BER), and Cramer-Rao (CR) sub-bound of the channel estimation algorithm were obtained analytically, and their performances according to the signal-to-noise ratio were analyzed with computer simulations. From the obtained MSE, BER, and CR curves, it is concluded that the MSE performance of the proposed channel estimation algorithm is very high and the BER performance of the system obtained with the predicted channel information is very close to the BER performance when the channel is known perfectly.