A Low-Complexity Framework for Enhanced Sensing and Communication in Terahertz ISAC Systems

This paper proposes an embedded pilot-assisted parameter estimation scheme for orthogonal time frequency space (EP-OTFS) in terahertz integrated sensing and communication (THz-ISAC) systems. In each OTFS frame, the pilot, guard, and data symbols are arranged in the delay-Doppler (DD) plane to suitably avoid interference between the pilot and data symbols at the receiver. These symbol arrangements are specifically designed for OTFS over multipath THz channels with both integer and fractional delay and Doppler shifts. Then a dual-stage sensing parameter method is introduced. This method integrates the sensing functionality into the OTFS waveform by leveraging the embedded pilot design. This design enables accurate sensing while reducing the computational complexity of the estimation process. The data detection subsequently leverages the estimated channel information through an iterative channel estimation and data detection algorithm with low computational complexity that employs an equalizer based on the conjugate gradient method. The proposed OTFS design integrates channel estimation, parameter sensing, and data detection into a single frame. Simulation results and complexity analysis show that the proposed EP-OTFS sensing scheme achieves millimeter-level precision in range and centimeter-per-second-level precision in velocity estimation with reduced computational complexity compared to the traditional OTFS schemes. Additionally, it exhibits a reduced bit error probability, even in the presence of fractional DD effects.