Search

International audience; In mobile communications, the most widely used multicarrier technique, namely OFDM, is known to exhibit performance limitations in high speed scenarios. FBMC is an alternative approach and, among FBMC techniques, the recently introduced FBMC-PAM scheme has emerged as the most robust to CFO. Therefore, it is chosen here as the challenger of OFDM for performance comparison in doubly dispersive channels. After presentation of the system model, multicarrier scheme and wireless channel, an analytical evaluation of the BER in the particular case of single path time-varying channel is provided for FBMC-PAM and validated through simulations. Then, simulation results are reported for both OFDM and FBMC-PAM, using the 3GPP multipath channel models. It appears that FBMC-PAM can outperform OFDM in highly time-varying frequency-selective channels.

We consider the problem of the downlink transceiver design when it needs to operate in an ultralarge analog bandwidth; more specifically, we propose to use the FBMC/OQAM system in a scenario where many other multicarrier transceivers are not able to operate. For such a case, we consider the problems of the equalizer implementation and of the design of its coefficients, and we define proper procedures for solving them by resorting to the MMSE approach. The simulation results show the capability of the proposed solutions to operate also on the most hostile wireless channels at the price of a reasonable increase of the receiver computational complexity.

This paper deals with the problem of blind carrier-frequency offset (CFO) estimation for filter bank multicarrier systems with pulse amplitude modulation (FBMC-PAM). In FBMC-PAM a sequence of real-valued symbols can be transmitted over 2 $M$ spectrally separated subcarriers operating at the same symbol rate as OFDM. This recently proposed system does not require a cyclic prefix and, in the multi-user context, it provides a high level of spectral separation among users. In this paper is derived the maximum likelihood (ML) blind CFO estimator for AWGN channel under the assumption of low SNR conditions. Since the FBMC-PAM signal is a noncircular random process, the obtained ML estimator exploits both the unconjugate and the conjugate correlation (or relation function). The performance of the derived CFO estimator is compared with that of the two estimators obtained by separately maximizing the contribution to the ML cost function exploiting the unconjugate correlation or the conjugate correlation, and with the modified Cramer-Rao bound.

Filter bank multicarrier (FBMC) techniques are emerging alternatives to the widely used orthogonal frequency division multiplexing (OFDM) schemes for applications such as machine type communications or cognitive radio. Recently an FBMC scheme based on a sine prototype filter with overlapping factor K = 2 termed FBMC-PAM has been introduced. It assures an interesting performance, a reduced system latency and is robust to carrier-frequency offset. After presentation of the system model, an analytical expression of the BER of the single-tap sub-channel equalizer for the considered FBMC-PAM system operating in frequency-selective Rayleigh fading channel is obtained and compared with simulation results. Moreover, a performance comparison with an approximate single-tap sub-channel equalizer and with the previously proposed FBMC-PAM-2M, is presented.

We consider one of the most interesting candidates for the 5G transceiver system, the windowed OFDM, and we carry out an evaluation of the spectral efficiency that can be achieved in the uplink of small but crowded cells by means of a time-frequency multiple-access scheme. Our analysis is first carried out in the absence of multiple-access interference in order to determine a performance upper bound and to gain more insights about the role of the system parameters. The main contribution of the paper is constituted by the evaluation of the achievable spectral efficiency as a function of the most important transceiver parameters, and above all by the number of subcarriers assigned to each active mobile terminal. In particular, we show that, unlike the downlink scenario, the spectral efficiency in multiple access is not optimized by choosing the cyclic prefix length to be just larger than the multipath delay spread, but increasing it much more, especially when the number of adjacent subcarriers assigned to each terminal is small.