A Portable Base Station Assisted Localization with Grid Bias Elimination

• Published in: 2023 IEEE Wireless Communications and Networking Conference (WCNC) pp. 1 - 6
• Main Authors: Li, Zhuyin, Zhu, Xu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.03.2023
• Subjects: Base stations; Classification algorithms; Delay effects; Estimation; grid bias elimination; intelligent security; joint angle delay estimation; Location awareness; multiple signal classification; single portable base station localization; Turning; Wireless communication
• ISSN: 1558-2612
• DOI: 10.1109/WCNC55385.2023.10118684

Localization has always been one of the key issues for security applications. As security systems are turning more intelligent together with the development of smart information technologies, critical requirements for wireless target localization have challenged the traditional positioning techniques, including flexibility, portability, deployment cost, computational efficiency, and estimation accuracy, to name a few. Although the widely- accepted classical algorithm, MUltiple SIgnal Classification (MUSIC), has been proven to be an effective tool for the space-time estimation, it can hardly satisfy localization requirements under such security scenarios due to the high complexity and the bias error led by grid searching. Thus, in this paper, we propose a Joint Angle and Delay Estimation (JADE)-based localization algorithm using only one single portable base station, which eliminates the grid bias with low computational complexity. First, a MUSIC-based coarse JADE approach is proposed; then, a Taylor-series-based refinement method is introduced to eliminate the grid bias; and finally, the target mobile station is localized by the estimated time delay and angle information. The performance is evaluated by numerical simulations under various conditions, compared with five different existing algorithms. Our proposed MT-2D algorithm is proven to achieve a better estimation accuracy for the time delay, angle and position with a relatively low computational cost.