Near optimum OSIC-based ML algorithm in a quantized space for LTE-A downlink physical layer

In this paper, we propose a scalable and implementation efficient OSIC-based ML algorithm in a quantized space with higher performance for MIMO detection, which can be applied to the LTE-A downlink physical layer system. It is characterized by dividing the overall OSIC detector into small dimension blocks to reduce complexity. The proposed algorithm utilizes ML algorithm in a quantized space to detect the first data streams and overcome error propagation problem. Then, it applies small dimension OSIC block to detect other data streams. The mathematical analysis is illustrated and derived. This paper shows BER performance of the proposed algorithm and compares its performance with other algorithms. This paper also presents the computational complexity to show that it gives lower complexity close to optimal ML algorithm. Simulation results show that the proposed algorithm provides a better performance and low BER values compared to OSIC algorithm. Finally, the proposed algorithm enhances the detection in LTE-A system and gives results close to optimum ML.