Spectrally efficient FDM: Spectrum saving technique for 5G?

Abstract

Spectrally efficient frequency division multiplexing (SEFDM) improves spectral efficiency relative to the well known orthogonal frequency division multiplexing (OFDM). Optimal detection of SEFDM, to recover signals corrupted by inter carrier interference (ICI), has major drawbacks in the exponential growth of detection complexity with the enlargement of system size and modulation level. This poses several challenges to SEFDM practical implementations. In this work, we present and compare practicable detection algorithms for both uncoded and coded SEFDM systems. In the case of the uncoded system, we discuss a multi-band architecture termed block-spectrally efficient frequency division multiplexing (B-SEFDM) which subdivides the signal spectrum into several blocks, allowing each block to be detected separately. The other system discussed in the paper utilizes convolutional coding with an appropriate receiver comprising a fast Fourier transform (FFT) based demodulation and detection working alongside a standard Bahl-Cocke-Jelinek-Raviv (BCJR) decoder. Mathematical modelling results show the suitability of the detector for use in large size non-orthogonal multicarrier systems. In the presence of multipath frequency selective channel, system modelling results show that this coded system with 1024 sub-carriers can save up to 45% of bandwidth compared to an otherwise equivalent OFDM.