Performance evaluation of adaptive equalizers in EMFB-based Multicarrier Modulation schemes

This work considers adaptive subchannel equalization in Exponentially-Modulated Filter Bank (EMFB) transmultiplexers (TMUX) when the channel is highly frequency selective. Using higher order sub-equalizers, improves the performance in this condition but convergence rate of adaptive algorithms decreases by increasing their order. Recently, Fine-Coarse (FC) algorithm was proposed for improving the convergence rate but evaluating the Bit Error Rate (BER) performance in realistic channels was addressed to future works. On the other hand, when the channel is highly frequency selective as it cannot be assumed flat over subcarrier bandwidth, the subequalizer input will be corrupted by colored spectrum. Therefore the use of Affine Projection Algorithm (APA) that has robust performance even for highly colored excitation signals is proposed in this paper for such situations. The spectral property of sub-equalizer inputs will be studied. FC and APA algorithms are used as subchannel equalizers and compared with Normalized Least Mean Square (NLMS) algorithm in a frequency selective fading channel. Simulation results show that APA has better performance.