Matrix Pencil-Based Channel Estimation Algorithm for OTFS in Fractional Doppler Channels

Orthogonal Time Frequency Space (OTFS) modulation is a promising modulation technology, which can meet the communication requirements of future high-speed mobile scenarios. In OTFS modulation, obtaining channel state information (CSI) is crucial for data detection. We show that the delay-doppler (DD) domain channel estimation problem can be transformed into an exponential sinusoid signal parameter estimation problem in delay-time (DT) domain using Inverse Discrete Fourier Transform (IDFT). Hence, we propose a Matrix Pencil (MP)-based channel estimation algorithm to achieve fractional Doppler channel estimation. This algorithm can accurately decompose coherent Doppler shifts without the need of Doppler search space in presence of inter-Doppler interference (IDI), and thus has lower computational complexity. Our simulation results verify that the normalized mean square error (NMSE) and bit error rate (BER) of the proposed algorithm can outperform the baseline algorithms.