SORIS: A Self-Organized Reconfigurable Intelligent Surface Architecture for Wireless Communications

In this paper, a new reconfigurable intelligent surface (RIS) hardware architecture, called self-organized RIS (SORIS), is proposed. The architecture incorporates a microcontroller connected to a single-antenna receiver operating at the same frequency as the RIS unit elements, operating either in transmission or reflection mode. The transmitting RIS elements enable the low latency estimation of both the incoming and outcoming channels at the microcontroller’s side. In addition, a machine learning approach for estimating the incoming and outcoming channels involving the remaining RIS elements operating in reflection mode is devised. Specifically, by appropriately selecting a small number of elements in transmission mode, and based on the channel reciprocity principle, the respective channel coefficients are first estimated, which are then fed to a low-complexity neural network that, leveraging spatial channel correlation over RIS elements, returns predictions of the channel coefficients referring to the rest of elements. In this way, the SORIS microcontroller acquires channel state information, and accordingly reconfigures the panel’s metamaterials to assist data communication between a transmitter and a receiver, without the need for separate connections with them. Moreover, the impact of channel estimation on the proposed solution, and a detailed complexity analysis for the used model, as well as a wiring density and control signaling analysis, is performed. The feasibility and efficacy of the proposed self-organized RIS design and operation are verified by Monte Carlo simulations, providing useful guidelines on the selection of the RIS elements for operating in transmission mode for initial channel estimation.

💡 Research Summary

The paper introduces a novel hardware architecture for reconfigurable intelligent surfaces (RIS) called Self‑Organized RIS (SORIS), which addresses the long‑standing challenge of acquiring instantaneous channel state information (CSI) at the RIS without relying on a dedicated feedback link. In SORIS, each RIS element can operate either in transmission or reflection mode, and a single‑antenna receiver is integrated into the RIS microcontroller. A small subset of elements is switched to transmission mode; these “transmitting elements” radiate pilot signals that are directly captured by the microcontroller’s receiver. By exploiting the short, deterministic link between the controller and the transmitting elements, the controller can estimate the BS‑RIS and UE‑RIS channels for those elements using simple division of the measured BS‑RIS‑controller cascaded channel. Reciprocity is assumed for the uplink, so the same estimates apply to the reverse direction.

The remaining (typically thousands of) passive elements operate in reflection mode only. Their CSI cannot be measured directly, so the authors propose a low‑complexity recurrent neural network (RNN) – specifically a gated‑recurrent‑unit (GRU) model – that takes as input the estimated channel coefficients of the transmitting elements together with their spatial coordinates and outputs predicted channel coefficients for all non‑transmitting elements. The model leverages spatial correlation across the RIS aperture; because neighboring elements experience similar propagation conditions, the RNN can accurately infer the missing channels with a modest number of trainable parameters (on the order of a few thousand), making real‑time execution feasible on ASIC or low‑power CMOS hardware.

A comprehensive complexity analysis shows that the channel‑estimation phase scales linearly with the number K of transmitting elements (O(K) operations), while the prediction phase scales linearly with the total number of elements N (O(N)). Power consumption is dramatically reduced compared with hybrid RIS designs that require a dedicated RF chain (LNA, down‑converter, ADC) per active element; SORIS needs only a single RF chain for the controller and a few switches for the transmitting elements. Consequently, wiring density and PCB area are minimized, which is crucial for large‑scale metasurfaces operating at mmWave or THz frequencies.

Monte‑Carlo simulations evaluate the impact of the number and placement of transmitting elements on CSI accuracy. Results indicate that even with as few as 5 % of the elements in transmission mode, the correlation coefficient between predicted and true channels exceeds 0.9, provided the transmitting elements are spatially spread to capture the underlying correlation structure. The simulations also incorporate realistic hardware impairments of the microcontroller’s receiver (IQ imbalance, phase noise, quantization error). Under these non‑ideal conditions, the average channel‑estimation error remains below 1 dB, confirming robustness.

The paper further discusses practical aspects such as control signaling overhead (only a few hundred bits are needed to convey element selection and phase‑shift commands) and the deterministic nature of the short controller‑to‑element link, which eliminates latency concerns that plague conventional RIS where CSI must be sent from the base station over a microsecond‑scale backhaul.

In summary, SORIS achieves self‑organization and self‑adaptation of the RIS by embedding real‑time CSI acquisition within the surface itself, eliminating the need for external feedback, reducing power and hardware complexity, and maintaining high estimation accuracy even in the presence of hardware imperfections. The authors suggest future work on dynamic selection of transmitting elements, multi‑user extensions, and hardware prototyping to validate the concept in real‑world 6G and beyond scenarios.