Design of passive-friendly RIS channel estimation and feedback protocols

Develop efficient channel estimation and feedback protocols for reconfigurable intelligent surfaces (RISs) that preserve passive operation without relying on power amplifiers, analog-to-digital converters, digital-to-analog converters, or onboard computation, and determine the associated performance-versus-complexity trade-offs to enable practical optimization of RIS reflection phases in wireless networks.

Background

RIS-assisted communications aim to keep the surfaces nearly passive, avoiding power-hungry components such as power amplifiers and ADC/DACs. This raises a fundamental challenge: RISs still need channel knowledge to optimize reflection phases, yet obtaining and conveying this information while remaining passive is difficult.

The paper outlines two potential approaches: embedding sparse low-power sensors on the RIS to sense channels and relay them via a gateway, or estimating optimal reflection phases using only the combined (product) channel without individual transmitter–RIS and RIS–receiver channel knowledge. However, the optimality of the latter approach versus full channel knowledge, and the broader design of protocols that balance passivity with performance, remain unresolved.