Machine Learning Based Indoor Localization using Visible Light Communication

Visible light communication (VLC) system has received significant attention and it has been penetrating numerous zones of daily life on account of its promising feature of data transmissions, in parallel to illumination. VLC is also a potential candidate for green data communication as its transmission is free from all hazards present in other wireless technologies such as Wi-Fi etc. This technology carries a number of unique features such as a license-free spectrum (384 THz-789 THz), high transmission speed, dual-use (illumination and data communication), ubiquitous, low-cost components, and stable channel conditions. However, there are certain challenges in the design of this novel technology such as an efficient channel access scheme, its use for accurate localization, duplex transmission system, dimming control and a suitable form of modulated signal that could drive the optical transmitter (LEDs) circuit. In this proposal, we design a physical system to investigate two of the above-mentioned challenges, i.e., a suitable channel access scheme to boost system capacity and its use for localization applications. Non-Orthogonal Multiple Access (NOMA) permits simultaneous data transmission with each user operating in the same band and at the same time where they are distinguished by their power levels or specific codes. Thus, the infusion of NOMA into the VLC system effectively enhances the spectral efficiency (SE) which is the key to its use for localization accuracy. The proposed VLC-NOMA will be investigated for localization application based on machine learning approach.