Simplified Implementation of QRD-M Algorithms for MIMO Wireless Communication System

For Multiple Input Multiple Output (MIMO) system employing Spatial Multiplexing (SM), Maximum Likelihood Detection (MLD) is computationally complex which makes it practically infeasible. Sphere decoding and tree search techniques can achieve near ML performance with reduced complexity. This paper presents a simple implementation of QR Decomposition with M Survivals (QRD-M) algorithm used for detection of Spatially Multiplexed data streams in MIMO wireless communication system. Using this algorithm, performance of simple MIMO and Multiple Input Multiple Output Orthogonal Frequency division Multiplexing (MIMO-OFDM) systems have been studied in terms of Bit Error Rates (BER). The algorithm is based on tree search technique. A Breadth First Search (BFS) technique is used to implement the algorithm. The complexity of the algorithm can be reduced by applying limited search at each level of the tree. A modified version of QRD-M algorithm is also proposed in which we keep on reducing the number of survivals. This scheme has improved computational complexity with slight degradation of performance.