Multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) technology has been used widely in many wireless communication systems. Signals will be distorted when they are transmitted in wireless channels. For the reason that wireless channel is time or location variant, we have to estimate the channel impulse response and use the channel state information to compensate the channel distortion. In order to estimate the state of the channel, let the known training symbols put in front of the data symbols and use training symbols to estimate channel response. A typical channel estimate for MIMO OFDM systems is treated as spatially uncorrelated. However in many realistic scenarios, the channel tends to be spatially correlated. Indeed, we have no prior knowledge of the channel spatial correlation. So consider the spatial correlation, the channel can estimate accurately. And it is important that how to combine spatial correlation and channel estimation to reduce the estimation error. In the paper we propose a iterative channel spatial correlation and channel estimation algorithm. At first, channel spatial correlation estimation is obtained by synchronize symbols. The receiver uses the estimated channel to help the detection/decision of data symbol. And then the channel estimation treats the detected signals as known data to perform a next stage channel estimation iteratively. By utilizing the iterative channel estimation and signal detection process we can reduce the estimation error caused by channel spatial correlation estimation. The accuracy of the channel estimation can be improved by increasing the number of iteration process. Simulation results demonstrate the iterative spatial correlation and channel estimation algorithm can provide better mean-square-error performance.