Non-Invasive Self-Adaptive Information States’ Acquisition inside Dynamic Scattering Spaces
Ruifeng Li, Ruifeng Li, Jinyan Ma +6 more
Abstract Pushing the information states acquisition efficiency has been a long-held goal to reach the measurement precision limit inside scattering spaces. Recent studies point out that the maximum information states can be achieved by engineered modes with specific matrices and operators, however, they typically necessitate time-consuming iterations or partial intrusion for multi-targets. Whereas non-invasive designs have been substantially explored across various physical scenarios, acquiring information states non-invasively inside dynamic scattering spaces remains challenging as it involves intractable non-unique mapping problem. Here, we first demonstrate that non-invasive information states acquisition is feasible and propose a tandem-generated adversarial network framework that can perform self-adaptive wavefield control inside dynamic scattering spaces. As an example, we have carried out a proof-of-principle experiment in the microwave regime, and prove that by using only the external scattering matrix of the system, information states acquisition for multi-targets can achieve the Fisher information limit efficiently. Our work demonstrates the unprecedented potential for real-time, efficient, and complex wave manipulation, providing explicable perspectives for precise measurements inside dynamic complex systems, such as medical micro-manipulation, optical Internet of Things, and intelligent communications.
Research•Jan 1, 2024•8 citations