Indoor Electronic Shelf Label Localization in High-Density Wireless Network

This paper introduces a novel fingerprint-based localization algorithm tailored for low-power Electronic Shelf Labels (ESLs) in complex retail environments characterized by high-density wireless networks. By integrating weighted signal strength analysis with maximum likelihood estimation, the algorithm achieves significant improvements in localization accuracy. Experimental evaluations demonstrate the proposed method outperforms traditional techniques, achieving 71.63% exact shelf-level accuracy, 89.47% one-deviation shelf accuracy and 90.20% aisle-side accuracy. These results remain robust even with varying proportions of known ESLs and under diverse placement configurations, where shelf accuracy ranges from 68.04% to 74.09% as the proportion of known ESLs varies between 10% to 50%. Furthermore, a calibration step enhances the algorithm’s adaptability, increasing shelf-level accuracy by approximately 1–2%, thus ensuring consistent performance across real-world deployment scenarios. The proposed method offers a scalable, efficient, and adaptive solution, significantly advancing ESL-based localization in smart retail applications.