Pulse Shaping for Faster-than-Nyquist to Enable Low-Complexity Detection

In this paper, we consider the problem of designing pulse shaping for faster-than-Nyquist (FTN) signaling. The proposed pulse shape is based on the optimization constraint such that the resulting intersymbol interference (ISI) matrix possesses super increasing sequence criteria. The low-complexity symbol-by-symbol sequence estimator which relies on this specific criteria can perform reasonably well even for lower values of time packing
$\tau=0.6$
. We formulate the problem as a finite impulse response (FIR) design and propose a second-order code program (SOCP) based solution. Simulation results show that with our proposed pulse shaping design for
$\tau=\{0.8,0.6,0.5\}$
, we obtain 2 dB or more performance improvement at a bit-error-rate (BER) of 10
^–3
compared to the state-of-the-art existing detection schemes over an additive white Gaussian noise (AWGN) channel.