Fairness-Based Row Selection in Randomized Kaczmarz Precoding for Massive MIMO Systems

This paper addresses the challenge of computational complexity in extremely large-scale massive Multiple-Input Multiple-Output (MIMO) systems by proposing an innovative row selection strategy using the Sainte-Laguë (SL) method in the randomized Kaczmarz Algorithm. Previous studies have utilized methods such as the D’Hondt method, which, despite improving performance, exhibits biases toward larger resource allocations. The proposed SL approach, renowned for its fairness and proportionality, significantly mitigates these biases, ensuring more uniform updates across rows. Through extensive computer simulations under realistic non-stationary channel conditions with imperfect Channel State Information, the proposed method demonstrates superior bit error rate (BER) performance compared to conventional methods, particularly in scenarios with fewer iterations. Specifically, results show that the SL method rapidly converges to near-optimal performance, reducing computational complexity compared to the RZF method and enabling efficient real-time signal processing in massive MIMO systems. These findings underline the potential of this method to enhance computational efficiency and BER performance, offering a promising solution for next-generation wireless communication networks.