ADAPTIVE PSO-ENHANCED CHANNEL ESTIMATION FOR MIMO-OFDM SYSTEMS UNDER DIVERSE MODULATION AND ANTENNA CONFIGURATIONS

Abstract

This study presents an adaptive channel estimation (CE) framework for MIMO-OFDM systems that integrates Particle Swarm Optimization (PSO) with traditional Least Squares (LS) and Linear Minimum Mean Square Error (LMMSE) estimators to enhance estimation accuracy and robustness. The proposed system employs different configurations for transmit and receive antennas over an OFDM framework for CE. A PSO-based refinement stage is introduced to optimize the LMMSE estimate by minimizing the Mean Square Error (MSE) iteratively within bounded search regions. Simulation results across varying SNR levels and MIMO configurations demonstrate that the PSO-enhanced estimator significantly outperforms conventional LS and LMMSE methods in terms of MSE reduction and convergence stability. The performance is further validated across multiple modulation schemes (BPSK, 4-QAM, 16-QAM, 4-PSK, and 8-PSK), showing consistent gains, particularly under low-SNR and high-modulation conditions. Although larger MIMO configurations exhibit slightly higher error floors, the PSO-based approach maintains strong adaptability and precision. Overall, the proposed LMMSE-PSO hybrid framework provides a scalable, noise-resilient, and computationally efficient solution for accurate CE in next-generation MIMO-OFDM communication systems.