IRS-Assisted OTFS System: Design and Analysis

Abstract

Intelligent reflecting surface (IRS) is a recent technology garnering lots of attention amongst the wireless communication research community due to its ability to improve the channel conditions extravagantly. The main feature of IRS technology is its capability to tune the phases of the reflecting elements to trigger a cohesive amalgamation of various multipath. IRS can easily co-exist with almost all technologies, thereby offering flexibility to the designers to come up with attractive systems. Further, orthogonal time frequency space (OTFS) is a recent 2D modulation technique that can mitigate the drawbacks of the traditional orthogonal frequency division multiplexing (OFDM) for fifth generation (5G) and beyond networks, especially in high Doppler wireless environments like vehicle-to-everything (V2X) and millimeter wave (mmWave) communications. This paper presents the system architecture of the IRS-assisted OTFS scheme in the delay-Doppler (DD) domain, including the modulation, demodulation, and detection methods. The selection of the reflection coefficients play the most vital role in the IRS-OTFS system, like any IRS system. IRS reflection optimization cannot perfectly line up all multipath with the same phase in the frequency-selective channels. We adopt an IRS phase optimization method where only the strongest delay-Doppler path is phase-aligned, while the paths with lesser strengths will be out of phase with the direct path. This reconfigurability of IRS elements enhances the OTFS modulation scheme, thereby significantly mitigating the multipath fading effects in high Doppler channels. Additionally, we propose a channel estimation for the IRS-OTFS system where both data and pilots are embedded in the same OTFS frame with minimal guard-band overhead. The paper presents extensive simulation results on the error-rate performance of the proposed system over Rayleigh and Rician fading channels. The effects of the number of IRS elements, the phase selections, and the channel estimation method on the system's performance are investigated. The comprehensive simulation study demonstrates that by exploiting the features of both OTFS and IRS, the proposed approach can provide impressive performance in high Doppler environments.