Multi-sphere decoding of block segmented SEFDM signals with large number of sub-carriers and high modulation order

Abstract

A non-orthogonal multicarrier signal, termed spectrally efficient frequency division multiplexing (SEFDM), is investigated in this work. It improves spectral efficiency by compressing sub-carrier spacing below the symbol rate at the cost of self-created inter carrier interference (ICI). Sphere decoding (SD) is an efficient method to recover signals approaching maximum likelihood (ML) performance. However, the complexity of the SD approaches that of ML with the increase of system size. Studies in this work show that for a small number of sub-carriers, SEFDM signals with low order modulation formats outperform spectral efficiency equivalent OFDM signals modulated by higher order modulation symbols. A key achievement is that 16QAM SEFDM signal outperforms 64QAM OFDM signal at high Eb/N <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> of the same spectral efficiency. In order to maintain the performance benefit and reduce the complexity of SD for large size SEFDM signals, multi-sphere decoding for a multi-block architecture is applied. For a large number of sub-carriers, the multi-sphere architecture works well for SEFDM signals modulated by 4QAM symbols. Whilst for 16QAM symbols, the performance is related to the number of sub-carriers. This work offers an efficient detection solution for SEFDM signals and reveals challenges. It paves the way for future study of signal detection of large size interfered multicarrier signals.