Data-Aided Fractional Delay-Doppler Channel Estimation with Embedded Pilot Frames in DZT-Based OTFS

Orthogonal time frequency space (OTFS) implementation using discrete Zak transform (DZT) has performance and complexity advantages in channels with high Doppler spreads. In this paper, we propose a data-aided approach to channel estimation in DZT-based OTFS systems with fractional delays and Dopplers. The proposed method employs an embedded pilot frame (consisting of pilot and data symbols) for channel estimation and leverages detected data to enhance channel estimation accuracy. The algorithm adopts a path-by-path approach where the channel parameters for one path are estimated before proceeding to the next, without assuming the knowledge of the number of paths. Considering the embedded nature of the pilot and data symbols in a frame, the effect of data signal-to-noise ratio (SNR) on the channel estimation performance and the effect of pilot SNR on the data detection performance are evaluated, leading to the data-aided approach to achieve improved performance. Simulation results show that the proposed estimation algorithm achieves significantly better performance compared to threshold based estimation algorithm.