Iterative Receiver for Non-Orthogonal Waveforms Based on the Sum-Product Algorithm

TL;DR

This paper introduces an iterative receiver for non-orthogonal waveforms using the Sum-Product algorithm, demonstrating effective BER performance and reduced complexity for GFDM and FBMC-OQAM systems under various channel conditions.

Contribution

It presents a novel graphical representation and iterative receiver algorithm for non-orthogonal waveforms based on SPA, with performance and complexity analysis.

Findings

BER curves closely match theoretical predictions

Proposed scheme reduces computational complexity

Effective in AWGN and Rayleigh fading channels

Abstract

Based on the application of the Sum-Product algorithm (SPA) over factor graphs, this paper presents a graphical representation of generalized frequency division multiplexing (GFDM) and filter bank multicarrier with offset QAM (FBMC-OQAM). FBMC-OQAM was chosen because it has the advantage of reducing the algorithm's complexity, since it is directly related to the number of possible values assumed by the transmitted data symbols. The receiver algorithm performance is evaluated by the bit error ratio (BER) estimation considering two channel models, additive white Gaussian noise (AWGN) and flat-fading time-variant (Rayleigh). Likewise, a computational complexity analysis is presented. Numerical results show that the BER curves of the proposed scheme present a good match compared with theoretical bit error probability curves.