Novel MIMO Detection With Improved Complexity for Near-ML Detection in MIMO-OFDM Systems

Abstract

In this paper, a novel efficient algorithm for MIMO detection in MIMO-OFDM systems has been proposed. This paper also shows that the complexity can be greatly reduced by applying the threshold in the breadth-first tree search algorithm step to eliminating paths. That is, this algorithm derives thresholds based on the LR-aided nonlinear algorithm and performs a simple tree search detection algorithm. This algorithm requires a lower computational complexity than the conventional QRD- $M$ detection algorithm and achieves the same bit error performance. After the novel detection algorithm is proposed, this paper applies this algorithm to the MIMO-OFDM system for performance comparison in terms of error performance and complexity. In many cases of MIMO-OFDM receivers, the proposed method will be an excellent option for implementation.