Wireless OFDM-OQAM with a Small Number of Subcarriers

Abstract

Orthogonal frequency division multiplexing based on offset quadrature amplitude modulation (OFDM-OQAM) is a multicarrier signaling technique which trades off robustness for spectral efficiency when compared to conventional OFDM with a cyclic prefix. In this paper, a novel matrix model for passband OFDM-OQAM signaling with a small number of subcarriers over a multipath frequency selective fading channel is presented. Specifically, in OFDM-OQAM a frequency selective channel is divided into many smaller but still frequency selective overlapping channels, so approximating the frequency response of a subchannel by the channel frequency response sampled at the subcarrier frequency may be inadequate. Channel effects may be better characterized if the frequency response of each subchannel is represented as a Taylor expansion at the subcarrier frequency. Simulation results show how the matrix model implemented by means of this approximation is a suitable model for OFDM-OQAM with a small number of subcarriers

Channel estimation in filter bank-based multicarrier systems: Challenges and solutions

“…The motivation behind this simplification is that it allows one to address the problem in a way similar to OFDM. However, and especially in propagation conditions envisaged in future networks, this assumption may be often quite inaccurate, due to the high frequency selectivity of the channel [8] and/or the small number of subcarriers employed to cope with frequency dispersion in fast fading environments [9]. In such cases, relying on the above assumption results in severe error floors at medium to high signal-to-noise ratio (SNR) values, which cancel the advantage of the FBMC modulation over OFDM [7].…”

Adaptive BLAST decision-feedback equalizer for MIMO-FBMC/OQAM systems

“…1 However, the previously mentioned advantages of FBMC/OQAM come at the cost of subcarrier functions being now orthogonal only in the real field, which means that there is always an intrinsic imaginary interference among (adjacent) subcarriers [11]. Moreover, the subchannels cannot, in general, be modeled as frequency flat, especially in timevarying environments where the number of subcarriers has to be small enough [8]. This makes signal processing tasks for FBMC/OQAM systems, such as channel equalization, more challenging, compared to OFDM/QAM.…”

MUSIC-based modal channel estimation for wideband OFDM-OQAM

“…We present in this paper a modal channel estimator based on the model [7] for the OFDM-OQAM received signal with wideband, frequency-selective subchannels. With the term modal we mean explicit estimation of the channel parameters in terms of delays and gains of each path.…”

“…We underline that the presented work is not an extension of the approach presented in [12] for time-delay estimation in OFDM with a cyclic prefix, since the OFDM-OQAM model [7] is not trivially suitable for application of the MUSIC algorithm. In fact, in OFDM with a cyclic prefix the channel is observable through noisy samples of the channel transfer function, while in wideband OFDM-OQAM the channel transfer function is only indirectly observable through its Taylor expansion around each subcarrier center frequency [7].…”

“…Section II briefly reviews the MUSIC estimator and the application condition of the algorithm with correlated additive noise is presented. In section III the OFDM-OQAM reference system model [7] is expressed in the form required by the MUSIC algorithm to estimate the path delays. Section IV describes the LS algorithm to estimate the path gains.…”