e-space
Manchester Metropolitan University's Research Repository

Performance Limits of Visible Light-Based Positioning for Internet-of-Vehicles: Time-Domain Localization Cooperation Gain

Zhou, Bingpeng and Liu, An and Lau, Vincent and Wen, Jinming and Mumtaz, Shahid and Bashir, Ali Kashif and Ahmed, Syed Hassan (2020) Performance Limits of Visible Light-Based Positioning for Internet-of-Vehicles: Time-Domain Localization Cooperation Gain. IEEE Transactions on Intelligent Transportation Systems. pp. 1-15. ISSN 1524-9050

[img]
Preview

Download (1MB) | Preview

Abstract

In this paper, we aim to give a unified performance limit analysis of the visible light-based positioning (VLP) for a vehicular user equipment (UE), which will help to understand the essence of time-domain localization cooperation and gain insights into how to improve the performance limit of the vehicular VLP system. This is challenging due to the complex system models and the complex dependency between UE location performance and orientation performance. To achieve the above goal, we will first characterize the closed-form error bounds of the UE location and orientation at each time slot, respectively, in terms of Fisher information. Generally, the VLP error will propagate over time as the vehicular UE moves, and hence the VLP error at the current time slot is affected by the VLP performance at the previous time slot, the UE mobility and the channel quality. Based on the obtained VLP error bounds, we then reveal the impact of prior UE location knowledge, UE mobility and signal-to-noise-ratio on the VLP performance. Furthermore, the time-domain evolution of the VLP error is studied, where the convergence of the time-domain VLP error evolution is established and its closed-form stable state is quantified, which will shed light on the long-term performance of the vehicular VLP system.

Impact and Reach

Statistics

Downloads
Activity Overview
91Downloads
73Hits

Additional statistics for this dataset are available via IRStats2.

Altmetric

Actions (login required)

Edit Item Edit Item