Advancements in Adaptive Equalization: A Comparative Study of the Recursive Non-Quadratic Algorithm and Traditional Techniques

This study evaluates the performance of the Recursive Non-Quadratic (RNQ) algorithm for adaptive equalization in digital communication systems, comparing it with the Normalized Least Mean Squares (NLMS) algorithm. Using simulations with 16-Phase Shift Keying (16-PSK) and 16-Quadrature Amplitude Modulation (16-QAM) signals, transmitted through a Rayleigh multi-path channel with Additive White Gaussian Noise (AWGN) at various Signal-to-Noise Ratios (SNRs), the RNQ algorithm was found to outperform NLMS. RNQ demonstrated superior symbol recovery accuracy, a more open eye diagram, and lower Mean Square Error (MSE), converging faster to the optimal solution. High roll-off factors and smaller filter sizes further enhanced RNQ's performance. These results affirm the RNQ algorithm's effectiveness in improving communication system performance and suggest future research directions.