An iterative pilot-aided channel estimation technique for space-frequency block coded (SFBC) multiple-input multiple-output orthogonal frequency division multiplexing systems is proposed. Traditionally, when channel estimation techniques are utilised, the SFBC information signals are decoded one block at a time. In the proposed algorithm, multiple blocks of SFBC information signals are decoded simultaneously. The proposed channel estimation method can thus significantly reduce the amount of time required to decode information signals compared to similar channel estimation methods proposed in the literature. The proposed method is based on the maximum likelihood approach that offers linearity and simplicity of implementation. An expression for the pairwise error probability (PEP) is derived based on the estimated channel. The derived PEP is then used to determine the optimal power allocation for the pilot sequence. The performance of the proposed algorithm is demonstrated in high frequency selective channels, for different number of pilot symbols, using different modulation schemes. The algorithm is also tested under different levels of Doppler shift and for different number of transmit and receive antennas. The results show that the proposed scheme minimises the error margin between slow and high speed receivers compared to similar channel estimation methods in the literature.