A Complexity-Reduced QRD-SIC Detector for Interleaved OTFS

Signal detectors are quite important to attain the diversity of doubly-dispersive wireless channels. Detectors based on message-passing (MP) of factor graphs have been regarded as the way to achieve the near-optimal performance for OTFS. In this paper, by deriving the pattern of the multipath vectorized channel matrix of the orthogonal time frequency space (OTFS) system, it is shown that short girth (i.e. girth-4) may exist in the Tanner graphs, which will degrade the performance of MP detectors, especially with high modulation orders. By introducing interleavers at the transmitter and receiver, the vectorized channel matrix turns out to be a sparse upper block Heisenberg matrix, whose structure is beneficial for the computation of matrix QR decomposition (QRD). Successive interference canceling (SIC) detectors based on QRD and sorted QRD are constructed to eliminate the cross-symbol interference and improve the reliability of the symbol-level channel. Simulation results show that for 4QAM, the QRD-based SIC detectors can achieve about 4dB gain at 10(-2) over the non-SIC detectors, while the sorted QRD-based SIC detectors can bring an additional 2dB at 10(-3), which is only 1dB gap from the MP. For 16QAM, the sorted SIC detectors show superior BER performance than the MP method, and for 64QAM, the MP detector reaches the error floor while SIC detectors show their excellent performance in all configurations.

Automated summary

This paper investigates the performance issues of message-passing (MP) detectors based on factor graphs in the orthogonal time frequency space (OTFS) system. It is found that there may exist short girths (i.e. girth-4) in the Tanner graphs, which degrade the performance of MP detectors, especially with high modulation orders. By introducing interleavers, the vectorized channel matrix becomes a sparse upper block Heisenberg matrix, which benefits the computation of matrix QR decomposition (QRD).