image: Figure | The schematic diagram of the programmable topological metasurface (PTM)-enabled integrated sensing and communication (ISAC) platform. An FPGA-controlled PTM is employed to manipulate surface wave propagation paths, enabling the collection of multidimensional EM data to train a neural network, hence realizing the human position sensing. Once the human position is recognized by the well-trained neural network, the PTM is reconfigured to regulate spatial waves, thereby enabling wireless communication. The nontrivial mode is realized by symmetry-broken valley-Hall domain-wall configurations for robust surface-wave sensing, whereas the trivial mode is realized by radiative phase-coding configurations for directional spatial-wave communication.
Credit: Din Ping Tsai et al.
Future wireless networks are expected not only to transmit information but also to perceive and understand their surrounding environment. This convergence of sensing and communication, known as integrated sensing and communication (ISAC), has emerged as a central objective for next-generation intelligent electromagnetic systems. However, conventional ISAC architectures often rely on separate sensing and communication modules, increasing hardware complexity, power consumption, and system cost. Achieving multiple electromagnetic functions on a unified platform remains a major challenge.
In a new paper published in Light: Science & Applications, a team of scientists, led by Professor Tie Jun Cui from the State Key Laboratory of Millimeter Waves, Southeast University, China, and Professor Din Ping Tsai from the Department of Electrical Engineering, City University of Hong Kong, Kowloon, China and co-workers have developed a trivial–nontrivial programmable topological metasurface that enables sensing and communication within a single reconfigurable electromagnetic platform. By dynamically switching between topological and radiative electromagnetic states, the metasurface supports both surface-wave-based human-position sensing and spatial-wave-based wireless communication. The proposed system employs a programmable topological metasurface controlled by an FPGA. In the nontrivial state, robust topological surface waves are guided along programmable domain-wall channels to collect multidimensional electromagnetic signatures associated with different human positions. These data are subsequently processed by a convolutional neural network to achieve high-accuracy position recognition. Once the target position is identified, the metasurface is reconfigured into a trivial radiative state, where it manipulates spatial electromagnetic waves to establish directional wireless communication links. Experimental results demonstrate a positioning accuracy exceeding 99% while simultaneously supporting reliable wireless information transmission. The transition between sensing and communication is achieved entirely through electronic reconfiguration, without mechanical movement or additional hardware subsystems.
These scientists summarize the significance of their work:
“We introduce a programmable topological metasurface that can switch between nontrivial and trivial electromagnetic states in real-time. This allows the same platform to perform two fundamentally different functions: topological surface-wave sensing and spatial-wave communication.”
“Topological electromagnetic states are usually studied for their robust wave-guiding properties. Here, we show that programmable topology can also serve as a system-level resource for integrated sensing and communication.”
“By combining programmable metasurfaces, topological physics, and machine learning, the proposed platform provides a new route toward multifunctional electromagnetic systems for future intelligent wireless environments.”
“The reported approach may find applications in future 6G networks, intelligent sensing systems, autonomous platforms, and adaptive electromagnetic environments, where sensing and communication are expected to operate in a deeply integrated manner.” the scientists forecast.
Journal
Light: Science & Applications
Article Title
Trivial-Nontrivial Programmable Topological Metasurfaces for Sensing and Communication