A reduced complexity receiver for multi-carrier faster-than-Nyquist signaling

Faster-than-Nyquist signaling originally proposed by J. E. Mazo promises higher spectral efficiencies by introducing controlled intersymbol interference at the transmitter at the cost of a higher receiver complexity. We propose a non-trellis based maximum likelihood (ML) receiver in conjunction with linear MMSE reception and interference cancellation for multi-carrier faster-than-Nyquist signaling (MC-FTN). The proposed receiver is able to collect the gains MC-FTN is promising while significantly reducing complexity compared to a receiver performing the optimum detection strategy (exhaustive search). We judge our receiver in terms of complexity, error rate performance and achieved spectral efficiency. The key question to be answered with respect to MC-FTN signaling in the view of a potential inclusion into a future 5G cellular communication system is: Are there scenarios in a realistic cellular communication setting for which gains from MC-FTN can be achieved at what complexity? This paper gives some insights covering all three aspects (scenarios, performance gains, complexity).