Traditional wireless relays and Intelligent Reflecting Surfaces (IRS) act like mirrors: they bounce signals to users on the same side as the transmitter. While effective for extending range, this "half-space" constraint creates a fundamental blind spot. If a user moves behind the surface—for instance, into a room separated by a window where the surface is deployed—the signal enhancement vanishes.
To achieve truly ubiquitous connectivity, researchers are developing a new class of "transflective" surfaces. These devices, often called Simultaneously Transmitting and Reflecting (STAR) RIS or Intelligent Omni-Surfaces (IOS), can manipulate electromagnetic waves in a full 360-degree radius by reflecting and refracting signals simultaneously.
From Opaque Mirrors to Smart Glass
The core intuition behind this shift is moving from an opaque reflective model to a semi-transparent, controllable one. As Shuhang Zhang, Hongliang Zhang, and Marco Di Renzo argue in their work on Intelligent Omni-Surfaces, a surface that can both reflect and transmit signals effectively doubles the service area of a single base station.
In a STAR-RIS, each meta-atom (the individual unit cell of the surface) is designed to split incident power. A portion of the energy is reflected back into the primary half-space, while the remainder is refracted through the surface to the "transmission" half-space. By electronically tuning the phase shifts of both components, the surface can perform beamforming for users on both sides of the wall at once.
Operating Protocols for Full-Space Control
Managing a surface that serves two sides simultaneously requires new resource allocation strategies. Chenyu Wu, Xidong Mu, and Yuanwei Liu have identified three primary protocols for STAR-RIS operation:
- Energy Splitting (ES): Every element splits its power between reflection and transmission. This offers the highest degrees of freedom but requires complex hardware.
- Mode Switching (MS): Each element is dedicated to either reflection or transmission at any given time. This simplifies the hardware but reduces the spatial resolution of the beams.
- Time Switching (TS): The entire surface alternates between reflection and transmission modes in rapid succession.
While Energy Splitting generally provides the best data rates, research by Chenyu Wu suggests that Time Switching is often superior for the initial, critical step of channel estimation.
The Hardware Reality of Coupled Phase Shifts
A significant challenge in this field is that transmission and reflection are not physically independent. In a passive, lossless meta-atom, the phase shift you apply to the reflected wave is mathematically tied to the phase shift of the transmitted wave.
Yuanwei Liu and Robert Schober introduced a coupled phase-shift model to account for these constraints. Ignoring this coupling leads to over-optimistic performance estimates. To address this, Zhaolin Wang developed optimization frameworks that maintain high performance even when the hardware cannot achieve "perfect" independent phase control.
Beyond 2D: Stacked and Interconnected Surfaces
The evolution of this technology is moving toward more complex 3D architectures and interconnected elements to overcome the limitations of simple "pixel-like" surfaces.
- Stacked Intelligent Metasurfaces (SIM): Instead of a single layer, Jiancheng An 0001, Mérouane Debbah, and Marco Di Renzo propose stacking multiple layers of metasurfaces. This creates a wave-domain processor that can perform complex signal processing—like interference nulling and multiuser beamfocusing—at the speed of light, potentially replacing power-hungry digital beamforming at the base station.
- Beyond Diagonal RIS (BD-RIS): Traditional surfaces are "diagonal," meaning each element acts independently. Bruno Clerckx and Hamad Yahya have pioneered BD-RIS, where elements are interconnected. This allows for more sophisticated scattering matrices, enabling the surface to perform tasks like Maximum Ratio Transmission (MRT) entirely in the passive domain.
Applications in 6G and Sensing
The ability to control the full 360-degree environment makes these surfaces ideal for next-generation access schemes. Integrating STAR-RIS with Non-Orthogonal Multiple Access (NOMA) is a recurring theme in the literature. Xinwei Yue and Yuanwei Liu have shown that STAR-RIS-NOMA can significantly outperform traditional orthogonal systems by exploiting the additional spatial degrees of freedom to separate users in both the reflection and transmission regions.
Furthermore, the technology is expanding into Integrated Sensing and Communications (ISAC). By embedding sensors into the STAR-RIS substrate, researchers like Zhaolin Wang are working toward full-space sensing, allowing the network to "see" and communicate with users in every direction simultaneously.