DMOA-DRIVEN CHANNEL ESTIMATION FOR OFDM: ROBUST PERFORMANCE ACROSS MODULATION ORDERS, PILOT DENSITIES AND 3GPP FADING MODELS

Abstract

This work proposes an innovative channel estimation technique for OFDM systems using the Dwarf Mongoose Optimization Algorithm (DMOA) without requiring any channel statistics. In this technique, the DMOA algorithm is used to search for the optimal parameters of the effective SNR, RMS delay spread, and Doppler frequency to minimize the pilot domain estimation error. Unlike the conventional MMSE method, which requires channel correlation matrices to be known in advance, this method adaptively estimates the parameters using the metaheuristic search algorithm. Simulation results prove that the proposed method consistently performs better than the conventional LS method in all modulation formats (QPSK, 16-QAM, 64-QAM, and 256-QAM), pilot densities (1/3, 1/6, 1/9, and 1/12), and 3GPP channel models (TDLC-300, EPA, EVA, and ETU). The results are particularly significant in the medium to high SNR region, in which the LS method shows significant error floor. In addition, the proposed method shows robust results in sparse pilot environments and effectively reduces the degradation caused by increasing pilot spacing. In all scenarios, DMOA achieves near-optimal results compared to the ideal MMSE method without requiring any statistical information. The proposed method also shows better results compared to PSO and DE in terms of stable convergence and reduced estimation error for the entire range of SNR. This shows DMOA to be a potential method for channel estimation in future OFDM systems.