In this paper, we investigate low complexity detection method to compensate the severe inter- symbol interference (ISI) introduced in Faster- than-Nyquist (FTN) Signaling based on a special circulated block transmission model. First, the optimal detection problem is formulated as a Boolean Quadratic Program (BQP). Through relax technique, the original non-convex problem is approximated by a L1-Norm constraint minimization problem. The relaxed convex problem is then efficiently tackled by the proposed Alternating Direction Method of Multipliers (ADMM) based algorithm. Finally, we generalize the algorithm to parallel computing implementation and High-order modulation case. Simulation results show that the proposed scheme can achieve better trade-off between Bit Error Rate (BER) performance and computational complexity than Frequency Domain Equalization (FDE) and Maximum Likelihood Sequence Estimation (MLSE).
Low-Complexity Symbol Detection for FTN Signaling by Combinatorial L1-Norm Relaxation
2018-08-01
191492 byte
Conference paper
Electronic Resource
English
Multiple-Symbol Differential Detection Of MPSK
NTRS | 1991
|Traffic Symbol Detection and Recognition System
Springer Verlag | 2021
|