An overview of optimal and sub-optimal detection techniques for a non orthogonal spectrally efficient FDM

Spectrally Efficient non orthogonal Frequency Division Multiplexing (SEFDM) Systems occupy less bandwidth than equivalent orthogonal FDM (OFDM). However, enhanced spectral efficiency comes at the expense of an increased complexity in the signal detection. In this work, we present an overview of different detection techniques that trade the error performance optimality for the signal recovery computational effort. Linear detection methods like Zero Forcing (ZF) and Minimum Mean Squared Error (MMSE) offer fixed complexity but suffer from a significant degradation of the Bit Error Rate (BER). On the other hand optimal receivers like Sphere Decoders (SD) achieve the optimal solution in terms of error performance. Notwithstanding, their applicability is severely constrained by the SEFDM signal dimension, the frequency separation between the carriers as well as the noise level in the system.