Visible Light Positioning using Received Signal Strength for Industrial Environments

Almadani, Yousef (2020) Visible Light Positioning using Received Signal Strength for Industrial Environments. Doctoral thesis (PhD), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (14MB) | Preview

Abstract

There is a forecast for exceptional digital data traffic growth due to the digitisation of industrial applications using the internet of things. As a result, a great need for high bandwidth and faster transmission data rates for future wireless networks has emerged. One of the considered communication technologies that can assist in satisfying this demand is visible light communications (VLC). VLC is an emerging technology that uses the visible light spectrum by mainly utilising lightemitting diodes (LEDs) for simultaneous indoor lighting and high bandwidth wireless communication. Some of the applications of VLC are to provide high data rate internet in homes, offices, campuses, hospitals, and several other areas. One of these promising areas of application is for industrial wireless communications. The research project will provide a review of VLC applications intended for industrial applications with an emphasis on visible light positioning (VLP). In this research work, a three-dimensional (3D) positioning algorithm for calculating the location of a photodiode (PD) is presented. It solely works on measured powers from different LED sources and does not require any prior knowledge of the receiver’s height unlike other works in the literature. The performance of the proposed VLP algorithm in terms of positioning error is evaluated using two different trilateration algorithms, the Cayley–Menger determinant (CMD) and the Linear Least Squares (LLS) trilateration algorithms. The evaluation considers different scenarios, with and without receiver tilt, and with multipath reflections. Simulation results show that the CMD algorithm is more accurate and outperforms the LLS trilateration positioning algorithm. Furthermore, the proposed method has been experimentally assessed under two different LED configurations, with different degrees of receiver tilt, and in the presence of a fully stocked storage rack to examine the effect of multipath reflections on the performance of VLP systems. It was observed from simulations and experimental investigations that the widely used square LED-configuration results in position ambiguities for 3D systems while a non-lattice layout, such as a star-shaped configuration, is much more accurate. An experimental accuracy with a 3D median error of 10.5 cm was achieved using the CMD algorithm in a 4 m × 4 m × 4.1 m area with a horizontal receiver. Adding receiver tilt of 5◦ and 10◦ increases the median error by an average of 29% and 110%, respectively. The effect of reflections from the i storage rack has also been thoroughly examined using the two mentioned trilateration algorithms and showed to increase the 3D median positioning error by an average of 69% in the experimental testbed for the areas close to the storage rack. These results highlight the degrading effect of multipath reflections on VLP systems and the necessity to consider it when evaluating these systems. As the primary consideration for positioning systems in industrial environments is for mobile robots, the encouraging results in this thesis can be further improved though the use of a sensor fusion method.