Optimization of FBMC Waveform by Designing NPR Prototype Filter with Improved Stopband Suppression

Abstract

In comparison with orthogonal frequency division multiplexing (OFDM), the filter bank multicarrier (FBMC) can reduce the out-of-band emissions through well-designed prototype filters, e.g., the isotropic orthogonal transform algorithm. Accordingly, this paper investigates a design method of nearly perfect reconstruction (NPR) prototype filter to effectively achieve both the filter flexibility and the FBMC frequency selectivity, in which the Nyquist condition can be linearly constrained by using the filter autocorrelation coefficients, and then, the spectral factorization aimed at minimum stopband energy is applied to retrieve the final prototype filter. The computer results show advantages over conventional ones in terms of sidelobe suppression, especially in the region near the transition band. Such advantages finally result in bit-error-rate (BER) superiority in FBMC simulations, which definitely confirms the effectiveness of the proposed design method for NPR prototype filter. Moreover, we also study the coefficient quantization and length reduction of the proposed NPR prototype filter. Fortunately, the quantized NPR filter can preserve its advantage on sidelobe suppression as well as BER. Besides, the proposed filter can produce BER results comparative to the PHYDYAS filter, although our filter length is much smaller than that of the PHYDYAS filter. Finally, these positive results demonstrate that the proposed method can benefit the FBMC applications in terms of flexible performance and simplified complexity.