Search

Multicarrier faster-than-Nyquist (MFTN) is a spectral efficient modulation scheme for future communication systems. However, since MFTN relies on the time interval packing of adjacent symbols/pulses and frequency spacing packing between adjacent subcarriers, the subcarriers are not orthogonal to each other anymore, and which results in the efficient implementation of MFTN signaling system a major challenge in practical applications. In this paper, we propose an efficient fast Fourier transform (FFT)-based implementation of MFTN signaling system. The proposed scheme is feasible when the subcarriers in MFTN are placed by a fraction of the orthogonal subcarrier spacing. By exploiting multiple parallel FFT blocks and combining the phase rotated versions of the FFT outputs, the MFTN signaling system can be implemented with substantially reduced computational complexity than conventional scheme, i.e., the direct implementation.

Multicarrier faster-than-Nyquist (MFTN) signaling is a spectral efficient transmission scheme for future communication systems. Similar as general multicarrier signals, the high peak-to-average power ratio (PAPR) is also one of the main drawbacks of MFTN signals. Unfortunately, the PAPR problem in MFTN signaling system has rarely been considered in the literatures. In this paper, we investigate the PAPR reduction for MFTN signals. First, we derive an inverse fast Fourier transform (IFFT)-based implementation of the MFTN transmitter. Based on which, a novel partial transmit sequence (PTS) scheme is proposed for the PAPR reduction of MFTN signals. The proposed PTS scheme relies on a two-stage optimization of the phase weighting factors for the parallel IFFT blocks as well as the phase weighting factors for the information symbols before the summation operation. It is shown that the proposed scheme substantially outperforms the direct employment of conventional Nyquist multicarrier transmission systems-based PTS scheme and also robust to the overlap-add operation introduced by pulse shaping.

Nonlinear companding transform is an effi- cient method to reduce the high peak-to-average power ratio (PAPR) of multicarrier transmission systems. However, the introduced companding noise greatly degrades the bit-error- rate (BER) performance of the companded multicarrier sys- tems. In this paper, a simplified but effective scheme of es- timation and cancellation of companding noise for the com- panded multicarrier transmission system is proposed. By ex- pressing the companded signals as the summation of original signals added with a companding noise component, and sub- tracting this estimated companding noise from the received signals, the BER performance of the overall system can be significantly improved. Simulation results well confirm the great advantages of the proposed scheme over other conven- tional decompanding or no decompanding schemes under various situations.

Faster-than-Nyquist (FTN) signalling, as a bandwidth-efficient modulation format, attracts many researchers’ attention. However, the symbol rates of FTN signal are increasing when the packing ratios decreases. Moreover, the FTN signal with small packing ratios requires a high-rate analogue-to-digital converter. Since the bandwidth of FTN signal remains the same, the samples taken at Nyquist sampling rate do not suffer information loss. However, the variable packing ratios induce extra memory-consumption when the FTN signal is sampled at Nyquist sampling rate. To reduce the sampling rate and the memory-consumption, a new receiver of FTN signal with oversampling is proposed. Simulation results show that the receiver can perform close to the convolution code line. At the same time, the sampling rate and the memory-consumption can be reduced by adjusting the oversampling factors according to the packing ratios.

Faster-than-Nyquist (FTN) signaling is a nonorthogonal transmission scheme, in which the pulses appear faster than the rate known from the Nyquist intersymbol interference (ISI)-free transmission criterion. Due to the violation of the Nyquist theorem, while a higher transmission rate is offered, ISI is introduced and the receiver complexity is greatly improved in the FTN signaling. The precoding technique can preprocess transmitted data before shaping filtering at the transmitter to remove ISI and relieve the receiver of heavy equalization or estimation procedures in the FTN signaling. In this paper, we investigate three linear precoding schemes, i.e., singular value decomposition (SVD), GTMH(G-to-minus-half) and Cholesky factorization (Cho) to reduce ISI and improve the performance of signal detection in FTN systems. We evaluate the performance of these precoding methods in terms of the bit error rate (BER), mutual information and power spectrum density (PSD). Numerous simulation results demonstrate that the GTMH and Cho precoded FTN have better BER and mutual information performance than the SVD scheme, but at the cost of broadening the spectrum of the signal.

Faster-than-Nyquist signaling is a promising waveform scheme in the fifth generation mobile communications for achieving higher spectrum efficiency. However, the FTN-caused intentional intersymbol interference will make the dectection process more complex. In this paper, a new precoding scheme is proposed for faster-than-Nyquist (FTN) signaling system aiming at reducing the burden of the receiver. Aided with the proposed frequency-domain precoding at the transmitter, the receiver can estimate the transmitted symbols by performing the symbol-by-symbol detection, which reduces the detection complexity dramatically. Moreover, the proposed precoding method can be applied into the high-order modulation directly, such as M-PSK or M-QAM. Further, the performance of proposed precoding scheme for FTN is verified and tested. Numerical results demonstrate that the proposed precoding scheme can not only realize symbol-by-symbol detection without bit-error-rate (BER) loss but also not broaden the spectrum of FTN signal at the transmitter.

Practical characteristic curve of the high power amplifier (HPA) below the saturation voltage is a nonlinear instead of a rigorous linear function, which indicates that the dominant small signal components in the conventional simplex clipping method will also suffer a nonlinear distortion and consequently, degrades the bit-error-rate (BER) performance of the receiver. In this paper, a novel cooperation scheme of clipping with companding is proposed to enhance the BER performance of multicarrier transmission system. By employing the companding function to compensate the nonlinear characteristics of HPA, the asymptotic undistorted small signals can be obtained after passing through the HPA. It is shown by simulation experiments that the BER as well as the out-of-band radiation performance can be greatly improved with only a slightly complexity increases.

A simple nonlinear companding transform which makes full use of the statistical characteristics of the original multi-carrier signals and combining them with the simplest clipping theory is proposed to reduce the peak-to-average power ratio (PAPR) of multi-carrier systems. By making a simple truncation to the Rayleigh-distributed original signals, the proposed scheme not only has low implementation complexity but also outperforms conventional companding schemes.

In this paper, a novel companding scheme is proposed to reduce the peak-to-average power ratio (PAPR) of lattice orthogonal frequency division multiplexing (LOFDM) system. By transforming the statistics of original signals into a specified distribution form, which is defined by a continuous sine function, this scheme can achieve a simple companding form as well as an improved PAPR and bit-error-rate (BER) performance. Moreover, by introducing the variable companding parameters in the desired probability density function (PDF), a great design flexibility in the companding form and an effective trade-off between the PAPR reduction and BER performance can be achieved to satisfy various system demands. The general formulas of the proposed scheme are derived and a theoretical analysis regarding the achievable transform gain and the selection criteria of companding parameters are also conducted. Simulation results show that the proposed scheme can substantially outperform the conventional