Cogent Machine Learning Algorithm for Indoor and Underwater Localization Using Visible Light Spectrum

From last few years indoor localization has become more popular for wireless devices. The major reason for its popularity is to access current location efficiently. On the other side visible light communication is attaining interest of researchers due to high growth of wireless communication and solid state lighting. This network structure can produce high accuracy for the resident positioning electromagnetic environment. The proposed approach is viable for both air and underwater communication based on the visible light spectrum. We evaluate the technique for other supervised machine learning algorithms to analyse an accuracy, error distance and computational time for indoor localization in visible light communication network. Experimental work was made by using star topology supported by visible personal area network system based simulator, which corresponds an attribute of PHY and MAC layer for IEEE 802.15.7 standard designed for short range optical wireless communication. The evaluation was carried out for an accuracy, error distance and computational time. The results show that the suggested methodology achieves overall computational accuracy and deliver an acceptable location estimation error.