Localization in Wireless Sensor Networks Using Visible Light in Non-Line of Sight Conditions

Abstract

In this paper, a novel method to localize indoor wireless sensor nodes using visible light in non-line of sight (NLOS) condition is proposed. The proposed method is able to identify NLOS condition in a sensor network and subsequently localize the sensor nodes. Since visible light is used for localization, the reflection points are first localized using time difference of arrival in a maximum likelihood framework. The location of the sensor nodes is then estimated using range and reflection angles which are themselves computed using novel geometric methods. Simulations and real field wireless sensor node deployments are then used to evaluate the performance of the proposed method. Experimental results indicate that the proposed method is able to localize sensor nodes with a high degree of accuracy when compared to conventional methods under NLOS conditions. The method also demonstrates reasonable robustness under sensor and ambient noise conditions.