Pilot Optimization for OFDM-based OTFS Systems over Doubly Selective Channels

For the orthogonal frequency division multiplexing-based orthogonal time-frequency-space (OFDM-based OTFS) systems, this paper studies the pilot optimization problem under the framework of distributed compressed sensing (DCS) and complex exponential basic extension model (CE-BEM). In order to reduce the inter-subcarrier interference (ICI) caused by the doubly selection (DS) channel and improve the accuracy of sparse channel estimation based on DCS, we propose pilot structure design and pilot pattern optimization schemes, respectively. First, we insert multiple sets of pilot clusters in the frequency domain OFDM symbols, where each set of pilot clusters contains an effective pilot and a guard pilot, and the data on both sides of the effective pilot is set equal to mitigate adjacent ICI. Then, through mutual coherence property, the optimization problem of effective pilot patterns in co-sparse symbols is transformed into a set of optimization problems of pilot interval, and a pilot pattern design algorithm based on pilot intervals is proposed. Finally, simulation results verify the effectiveness of the pilot structure design scheme, and the proposed algorithm not only has low computational complexity but also has superior system performance.