Machine Learning-Based Anchor-Aware Conditional Flow Matching for RF Localization in Wireless Capsule Endoscopy

Reliable RF localization for capsule endoscopy remains challenging due to complex in-body propagation and measurement uncertainty. In this work, we explore an anchor-aware flow matching approach to improve localization robustness while staying consistent with the anchor geometry and measurements.

Co-authors: Dr. Muhammad Qamar Satti, Dr. Kamil Yavuz Kapusuz, Dr. Mohamed Adhnan Thaha, Dr. Akram Alomainy

Research carried out at Queen Mary University of London, School of Electronic Engineering and Computer Science, Antennas and Electromagnetics Research Group and CE-Track.