Semi-blind Channel Estimation In MIMO-OFDM Wireless System

Accurate channel estimation is crucial for high data rates and high reliability of multiple input multiple output orthogonal frequency division multiplexing (MIMO-OFDM) wireless system. Usually it can be done by sending training data or pilot tones. The training or pilot tones overhead required, unfortunately, is a serious impact on the utilization of limited system bandwidth, especially in MIMO system. The blind channel estimation saves bandwidth and increases spectrum efficiency, but has high level of computation and estimation of uncertainty. Semi-blind estimation based subspace is an algorithm that adding few pilot tones in the blind channel estimation, solves uncertainty in blind channel estimation. But there is still the problem of high computational complexity, large of computing. In order to improve performance of algorithm, introducing the projection approximation subspace tracking (PAST) method for tracking subspace in semi-blind channel estimation algorithm based on the decomposition of subspace. The MIMO-OFDM semi-blind channel estimation algorithm based on PAST tracking is proposed. The proposed algorithm reduces the high computational complexity of the singular value decomposition, enhances algorithm operational efficiency greatly. But signal subspace based PAST tracking is unable to ensure the orthonormality of the weight matrix at each iteration, and probably tracks signal subspace inaccurately. To ensure tracking signal subspace accurately and each iteration process orthogonal signal, the orthogonal PAST (OPAST) tracking signal subspace methods is proposed. According to the orthonormality of the signal subspace and MIMO-OFDM system model, the estimated value of semi-blind channel impulse response space-based signals is achieved. Theoretical analysis and simulation results show that, compared to the original algorithm, the performance of improved algorithm maintained unchanged, while accelerates the convergence rate and reduces the complexity of computing. The proposed algorithm is simulated in the SUI-3 channel model of WIMAX, the simulation shows that the improved algorithm has good channel estimation performance.