Channel Estimation for Orthogonal Time Frequency Space Modulation using Recursive Least Squares

Abstract

For its capacity to provide high data rates to a wide number of users, 4G wireless communications had a huge success in the previous decade. With the Internet of Things (IoT) and high mobility scenarios such as vehicle-to-vehicle (V2X) connections on the horizon, the Orthogonal Time Frequency Space (OTFS) modulation scheme has ignited lot of attention in recent years as a viable alternative to OFDM, especially in scenarios involving high user mobility. OTFS has its specialty that it is designed in the delay-Doppler domain. OTFS modulation, when combined with an appropriate equaliser, easily leverages the whole channel variety in both time and frequency. It transforms a fading, time-varying wireless channel used by modulated communications like OFDM into a time-independent channel with a nearly complex channel gain for all symbols. This thesis makes a note on existing drawbacks of OFDM and highlights the usage of a new 2-D modulation scheme called OTFS modulation. It goes on to detail the various methods of channel estimate currently in use while installing OTFS and suggests the use of an adaptive algorithm for channel estimation in the delay-Doppler domain. The proposed algorithm, unlike widely used channel estimation methods, estimates channel gain in the time domain and Doppler taps in the delay-Doppler domain.