Abstract
In the MIMO-OFDM system channel estimation based on the superimposed pilot is one of most potential and powerful techniques for broad-band wireless communications. In most traditional algorithms, the channel parameters are estimated by calculating the first-order statistical properties of receiving information. However, there are some limitations in such methods. These methods require the first-order statistical property of the sending information is zero and the observed data is long enough, but it is difficult to meet the requirements in practice. In some cases, it even leads to a large channel estimation error. To solve this problem, a new judgment, iteration algorithm is proposed. This algorithm can weaken the effect of data transmission on the pilot frequency and reduce the estimation error. The simulation results indicate that the new algorithm exhibits good performance on channel utilization and the estimation accuracy without increasing the system complexity.